patch-3.7.1 (#1178)

* patch 3.7.1

* fix bug for 2 processors on image to image pipeline (#1177)

* reduce default value of target_frame_amount

* restore old performance

* restore old performance

* restore old performance

* restore old performance

.github/workflows/ci.yml

@@ -33,7 +33,7 @@ jobs:
     uses: actions/setup-python@v5
     with:
      python-version: '3.12'
-  - run: python install.py --onnxruntime default --skip-conda
+  - run: python install.py default --skip-conda
   - run: pip install pytest
   - run: pytest
  report:
@@ -48,7 +48,7 @@ jobs:
     uses: actions/setup-python@v5
     with:
      python-version: '3.12'
-  - run: python install.py --onnxruntime default --skip-conda
+  - run: python install.py default --skip-conda
   - run: pip install coveralls
   - run: pip install pytest
   - run: pip install pytest-cov

LICENSE.md

@@ -1,3 +1,3 @@
 OpenRAIL-AS license
 
-Copyright (c) 2025 Henry Ruhs
+Copyright (c) 2026 Henry Ruhs

facefusion.ini

@@ -32,6 +32,9 @@ reference_face_position =
 reference_face_distance =
 reference_frame_number =
 
+[face_tracker]
+face_tracker_score =
+
 [face_masker]
 face_occluder_model =
 face_parser_model =
@@ -50,6 +53,9 @@ trim_frame_end =
 temp_frame_format =
 keep_temp =
 
+[frame_distribution]
+target_frame_amount =
+
 [output_creation]
 output_image_quality =
 output_image_scale =
@@ -67,7 +73,8 @@ processors =
 age_modifier_model =
 age_modifier_direction =
 background_remover_model =
-background_remover_color =
+background_remover_fill_color =
+background_remover_despill_color =
 deep_swapper_model =
 deep_swapper_morph =
 expression_restorer_model =
@@ -124,7 +131,6 @@ execution_thread_count =
 
 [memory]
 video_memory_strategy =
-system_memory_limit =
 
 [misc]
 log_level =

facefusion.py

@@ -4,7 +4,8 @@ import os
 
 os.environ['OMP_NUM_THREADS'] = '1'
 
-from facefusion import core
+from facefusion import conda, core
 
 if __name__ == '__main__':
+	conda.setup()
 	core.cli()

facefusion/app_context.py

@@ -5,12 +5,14 @@ from facefusion.types import AppContext
 
 
 def detect_app_context() -> AppContext:
+	jobs_path = os.path.join('facefusion', 'jobs')
+	uis_path = os.path.join('facefusion', 'uis')
 	frame = sys._getframe(1)
 
 	while frame:
-		if os.path.join('facefusion', 'jobs') in frame.f_code.co_filename:
+		if jobs_path in frame.f_code.co_filename:
 			return 'cli'
-		if os.path.join('facefusion', 'uis') in frame.f_code.co_filename:
+		if uis_path in frame.f_code.co_filename:
 			return 'ui'
 		frame = frame.f_back
 	return 'cli'

facefusion/args.py

@@ -7,6 +7,85 @@ from facefusion.types import ApplyStateItem, Args
 from facefusion.vision import detect_video_fps
 
 
+def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
+	apply_state_item('command', args.get('command'))
+	apply_state_item('temp_path', args.get('temp_path'))
+	apply_state_item('jobs_path', args.get('jobs_path'))
+	apply_state_item('source_paths', args.get('source_paths'))
+	apply_state_item('target_path', args.get('target_path'))
+	apply_state_item('output_path', args.get('output_path'))
+	apply_state_item('source_pattern', args.get('source_pattern'))
+	apply_state_item('target_pattern', args.get('target_pattern'))
+	apply_state_item('output_pattern', args.get('output_pattern'))
+	apply_state_item('face_detector_model', args.get('face_detector_model'))
+	apply_state_item('face_detector_size', args.get('face_detector_size'))
+	apply_state_item('face_detector_margin', normalize_space(args.get('face_detector_margin')))
+	apply_state_item('face_detector_angles', args.get('face_detector_angles'))
+	apply_state_item('face_detector_score', args.get('face_detector_score'))
+	apply_state_item('face_landmarker_model', args.get('face_landmarker_model'))
+	apply_state_item('face_landmarker_score', args.get('face_landmarker_score'))
+	apply_state_item('face_selector_mode', args.get('face_selector_mode'))
+	apply_state_item('face_selector_order', args.get('face_selector_order'))
+	apply_state_item('face_selector_age_start', args.get('face_selector_age_start'))
+	apply_state_item('face_selector_age_end', args.get('face_selector_age_end'))
+	apply_state_item('face_selector_gender', args.get('face_selector_gender'))
+	apply_state_item('face_selector_race', args.get('face_selector_race'))
+	apply_state_item('reference_face_position', args.get('reference_face_position'))
+	apply_state_item('reference_face_distance', args.get('reference_face_distance'))
+	apply_state_item('reference_frame_number', args.get('reference_frame_number'))
+	apply_state_item('face_tracker_score', args.get('face_tracker_score'))
+	apply_state_item('face_occluder_model', args.get('face_occluder_model'))
+	apply_state_item('face_parser_model', args.get('face_parser_model'))
+	apply_state_item('face_mask_types', args.get('face_mask_types'))
+	apply_state_item('face_mask_areas', args.get('face_mask_areas'))
+	apply_state_item('face_mask_regions', args.get('face_mask_regions'))
+	apply_state_item('face_mask_blur', args.get('face_mask_blur'))
+	apply_state_item('face_mask_padding', normalize_space(args.get('face_mask_padding')))
+	apply_state_item('voice_extractor_model', args.get('voice_extractor_model'))
+	apply_state_item('trim_frame_start', args.get('trim_frame_start'))
+	apply_state_item('trim_frame_end', args.get('trim_frame_end'))
+	apply_state_item('temp_frame_format', args.get('temp_frame_format'))
+	apply_state_item('keep_temp', args.get('keep_temp'))
+	apply_state_item('target_frame_amount', args.get('target_frame_amount'))
+	apply_state_item('output_image_quality', args.get('output_image_quality'))
+	apply_state_item('output_image_scale', args.get('output_image_scale'))
+	apply_state_item('output_audio_encoder', args.get('output_audio_encoder'))
+	apply_state_item('output_audio_quality', args.get('output_audio_quality'))
+	apply_state_item('output_audio_volume', args.get('output_audio_volume'))
+	apply_state_item('output_video_encoder', args.get('output_video_encoder'))
+	apply_state_item('output_video_preset', args.get('output_video_preset'))
+	apply_state_item('output_video_quality', args.get('output_video_quality'))
+	apply_state_item('output_video_scale', args.get('output_video_scale'))
+
+	if args.get('output_video_fps') or is_video(args.get('target_path')):
+		output_video_fps = normalize_fps(args.get('output_video_fps')) or detect_video_fps(args.get('target_path'))
+		apply_state_item('output_video_fps', output_video_fps)
+
+	available_processors = [ get_file_name(file_path) for file_path in resolve_file_paths('facefusion/processors/modules') ]
+	apply_state_item('processors', args.get('processors'))
+
+	for processor_module in get_processors_modules(available_processors):
+		processor_module.apply_args(args, apply_state_item)
+
+	apply_state_item('open_browser', args.get('open_browser'))
+	apply_state_item('ui_layouts', args.get('ui_layouts'))
+	apply_state_item('ui_workflow', args.get('ui_workflow'))
+	apply_state_item('execution_device_ids', args.get('execution_device_ids'))
+	apply_state_item('execution_providers', args.get('execution_providers'))
+	apply_state_item('execution_thread_count', args.get('execution_thread_count'))
+	apply_state_item('download_providers', args.get('download_providers'))
+	apply_state_item('download_scope', args.get('download_scope'))
+	apply_state_item('benchmark_mode', args.get('benchmark_mode'))
+	apply_state_item('benchmark_resolutions', args.get('benchmark_resolutions'))
+	apply_state_item('benchmark_cycle_count', args.get('benchmark_cycle_count'))
+	apply_state_item('video_memory_strategy', args.get('video_memory_strategy'))
+	apply_state_item('log_level', args.get('log_level'))
+	apply_state_item('halt_on_error', args.get('halt_on_error'))
+	apply_state_item('job_id', args.get('job_id'))
+	apply_state_item('job_status', args.get('job_status'))
+	apply_state_item('step_index', args.get('step_index'))
+
+
 def reduce_step_args(args : Args) -> Args:
 	step_args =\
 	{
@@ -37,95 +116,3 @@ def collect_job_args() -> Args:
 		key: state_manager.get_item(key) for key in job_store.get_job_keys() #type:ignore[arg-type]
 	}
 	return job_args
-
-
-def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
-	# general
-	apply_state_item('command', args.get('command'))
-	# paths
-	apply_state_item('temp_path', args.get('temp_path'))
-	apply_state_item('jobs_path', args.get('jobs_path'))
-	apply_state_item('source_paths', args.get('source_paths'))
-	apply_state_item('target_path', args.get('target_path'))
-	apply_state_item('output_path', args.get('output_path'))
-	# patterns
-	apply_state_item('source_pattern', args.get('source_pattern'))
-	apply_state_item('target_pattern', args.get('target_pattern'))
-	apply_state_item('output_pattern', args.get('output_pattern'))
-	# face detector
-	apply_state_item('face_detector_model', args.get('face_detector_model'))
-	apply_state_item('face_detector_size', args.get('face_detector_size'))
-	apply_state_item('face_detector_margin', normalize_space(args.get('face_detector_margin')))
-	apply_state_item('face_detector_angles', args.get('face_detector_angles'))
-	apply_state_item('face_detector_score', args.get('face_detector_score'))
-	# face landmarker
-	apply_state_item('face_landmarker_model', args.get('face_landmarker_model'))
-	apply_state_item('face_landmarker_score', args.get('face_landmarker_score'))
-	# face selector
-	apply_state_item('face_selector_mode', args.get('face_selector_mode'))
-	apply_state_item('face_selector_order', args.get('face_selector_order'))
-	apply_state_item('face_selector_age_start', args.get('face_selector_age_start'))
-	apply_state_item('face_selector_age_end', args.get('face_selector_age_end'))
-	apply_state_item('face_selector_gender', args.get('face_selector_gender'))
-	apply_state_item('face_selector_race', args.get('face_selector_race'))
-	apply_state_item('reference_face_position', args.get('reference_face_position'))
-	apply_state_item('reference_face_distance', args.get('reference_face_distance'))
-	apply_state_item('reference_frame_number', args.get('reference_frame_number'))
-	# face masker
-	apply_state_item('face_occluder_model', args.get('face_occluder_model'))
-	apply_state_item('face_parser_model', args.get('face_parser_model'))
-	apply_state_item('face_mask_types', args.get('face_mask_types'))
-	apply_state_item('face_mask_areas', args.get('face_mask_areas'))
-	apply_state_item('face_mask_regions', args.get('face_mask_regions'))
-	apply_state_item('face_mask_blur', args.get('face_mask_blur'))
-	apply_state_item('face_mask_padding', normalize_space(args.get('face_mask_padding')))
-	# voice extractor
-	apply_state_item('voice_extractor_model', args.get('voice_extractor_model'))
-	# frame extraction
-	apply_state_item('trim_frame_start', args.get('trim_frame_start'))
-	apply_state_item('trim_frame_end', args.get('trim_frame_end'))
-	apply_state_item('temp_frame_format', args.get('temp_frame_format'))
-	apply_state_item('keep_temp', args.get('keep_temp'))
-	# output creation
-	apply_state_item('output_image_quality', args.get('output_image_quality'))
-	apply_state_item('output_image_scale', args.get('output_image_scale'))
-	apply_state_item('output_audio_encoder', args.get('output_audio_encoder'))
-	apply_state_item('output_audio_quality', args.get('output_audio_quality'))
-	apply_state_item('output_audio_volume', args.get('output_audio_volume'))
-	apply_state_item('output_video_encoder', args.get('output_video_encoder'))
-	apply_state_item('output_video_preset', args.get('output_video_preset'))
-	apply_state_item('output_video_quality', args.get('output_video_quality'))
-	apply_state_item('output_video_scale', args.get('output_video_scale'))
-	if args.get('output_video_fps') or is_video(args.get('target_path')):
-		output_video_fps = normalize_fps(args.get('output_video_fps')) or detect_video_fps(args.get('target_path'))
-		apply_state_item('output_video_fps', output_video_fps)
-	# processors
-	available_processors = [ get_file_name(file_path) for file_path in resolve_file_paths('facefusion/processors/modules') ]
-	apply_state_item('processors', args.get('processors'))
-	for processor_module in get_processors_modules(available_processors):
-		processor_module.apply_args(args, apply_state_item)
-	# uis
-	apply_state_item('open_browser', args.get('open_browser'))
-	apply_state_item('ui_layouts', args.get('ui_layouts'))
-	apply_state_item('ui_workflow', args.get('ui_workflow'))
-	# execution
-	apply_state_item('execution_device_ids', args.get('execution_device_ids'))
-	apply_state_item('execution_providers', args.get('execution_providers'))
-	apply_state_item('execution_thread_count', args.get('execution_thread_count'))
-	# download
-	apply_state_item('download_providers', args.get('download_providers'))
-	apply_state_item('download_scope', args.get('download_scope'))
-	# benchmark
-	apply_state_item('benchmark_mode', args.get('benchmark_mode'))
-	apply_state_item('benchmark_resolutions', args.get('benchmark_resolutions'))
-	apply_state_item('benchmark_cycle_count', args.get('benchmark_cycle_count'))
-	# memory
-	apply_state_item('video_memory_strategy', args.get('video_memory_strategy'))
-	apply_state_item('system_memory_limit', args.get('system_memory_limit'))
-	# misc
-	apply_state_item('log_level', args.get('log_level'))
-	apply_state_item('halt_on_error', args.get('halt_on_error'))
-	# jobs
-	apply_state_item('job_id', args.get('job_id'))
-	apply_state_item('job_status', args.get('job_status'))
-	apply_state_item('step_index', args.get('step_index'))

facefusion/benchmarker.py

@@ -9,7 +9,7 @@ import facefusion.choices
 from facefusion import content_analyser, core, state_manager
 from facefusion.cli_helper import render_table
 from facefusion.download import conditional_download, resolve_download_url
-from facefusion.face_store import clear_static_faces
+from facefusion.face_store import clear_faces
 from facefusion.filesystem import get_file_extension
 from facefusion.types import BenchmarkCycleSet
 from facefusion.vision import count_video_frame_total, detect_video_fps
@@ -64,7 +64,7 @@ def cycle(cycle_count : int) -> BenchmarkCycleSet:
 		if state_manager.get_item('benchmark_mode') == 'cold':
 			content_analyser.analyse_image.cache_clear()
 			content_analyser.analyse_video.cache_clear()
-			clear_static_faces()
+			clear_faces()
 
 		start_time = perf_counter()
 		core.conditional_process()
@@ -89,7 +89,7 @@ def cycle(cycle_count : int) -> BenchmarkCycleSet:
 
 def suggest_output_path(target_path : str) -> str:
 	target_file_extension = get_file_extension(target_path)
-	return os.path.join(tempfile.gettempdir(), hashlib.sha1().hexdigest()[:8] + target_file_extension)
+	return os.path.join(tempfile.gettempdir(), hashlib.sha1(target_path.encode()).hexdigest() + target_file_extension)
 
 
 def render() -> None:

facefusion/camera_manager.py

@@ -43,9 +43,9 @@ def detect_local_camera_ids(id_start : int, id_end : int) -> List[int]:
 	local_camera_ids = []
 
 	for camera_id in range(id_start, id_end):
-		cv2.setLogLevel(0)
+		cv2.utils.logging.setLogLevel(0)
 		camera_capture = get_local_camera_capture(camera_id)
-		cv2.setLogLevel(3)
+		cv2.utils.logging.setLogLevel(3)
 
 		if camera_capture and camera_capture.isOpened():
 			local_camera_ids.append(camera_id)

facefusion/choices.py

@@ -1,8 +1,8 @@
 import logging
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_float_range, create_int_range
-from facefusion.types import Angle, AudioEncoder, AudioFormat, AudioTypeSet, BenchmarkMode, BenchmarkResolution, BenchmarkSet, DownloadProvider, DownloadProviderSet, DownloadScope, EncoderSet, ExecutionProvider, ExecutionProviderSet, FaceDetectorModel, FaceDetectorSet, FaceLandmarkerModel, FaceMaskArea, FaceMaskAreaSet, FaceMaskRegion, FaceMaskRegionSet, FaceMaskType, FaceOccluderModel, FaceParserModel, FaceSelectorMode, FaceSelectorOrder, Gender, ImageFormat, ImageTypeSet, JobStatus, LogLevel, LogLevelSet, Race, Score, TempFrameFormat, UiWorkflow, VideoEncoder, VideoFormat, VideoMemoryStrategy, VideoPreset, VideoTypeSet, VoiceExtractorModel
+from facefusion.types import Angle, AudioEncoder, AudioFormat, AudioTypeSet, BenchmarkMode, BenchmarkResolution, BenchmarkSet, DownloadProvider, DownloadProviderSet, DownloadScope, EncoderSet, ExecutionProvider, ExecutionProviderSet, FaceDetectorModel, FaceDetectorSet, FaceLandmarkerModel, FaceMaskArea, FaceMaskAreaSet, FaceMaskRegion, FaceMaskRegionSet, FaceMaskType, FaceOccluderModel, FaceParserModel, FaceSelectorGender, FaceSelectorMode, FaceSelectorOrder, FaceSelectorRace, Gender, ImageFormat, ImageTypeSet, JobStatus, LogLevel, LogLevelSet, Race, Score, TempFrameFormat, UiWorkflow, VideoEncoder, VideoFormat, VideoMemoryStrategy, VideoPreset, VideoTypeSet, VoiceExtractorModel
 
 face_detector_set : FaceDetectorSet =\
 {
@@ -12,15 +12,17 @@ face_detector_set : FaceDetectorSet =\
 	'yolo_face': [ '640x640' ],
 	'yunet': [ '640x640' ]
 }
-face_detector_models : List[FaceDetectorModel] = list(face_detector_set.keys())
+face_detector_models : List[FaceDetectorModel] = list(get_args(FaceDetectorModel))
-face_landmarker_models : List[FaceLandmarkerModel] = [ 'many', '2dfan4', 'peppa_wutz' ]
+face_landmarker_models : List[FaceLandmarkerModel] = list(get_args(FaceLandmarkerModel))
-face_selector_modes : List[FaceSelectorMode] = [ 'many', 'one', 'reference' ]
+face_selector_modes : List[FaceSelectorMode] = list(get_args(FaceSelectorMode))
-face_selector_orders : List[FaceSelectorOrder] = [ 'left-right', 'right-left', 'top-bottom', 'bottom-top', 'small-large', 'large-small', 'best-worst', 'worst-best' ]
+face_selector_orders : List[FaceSelectorOrder] = list(get_args(FaceSelectorOrder))
-face_selector_genders : List[Gender] = [ 'female', 'male' ]
+genders : List[Gender] = list(get_args(Gender))
-face_selector_races : List[Race] = [ 'white', 'black', 'latino', 'asian', 'indian', 'arabic' ]
+races : List[Race] = list(get_args(Race))
-face_occluder_models : List[FaceOccluderModel] = [ 'many', 'xseg_1', 'xseg_2', 'xseg_3' ]
+face_selector_genders : List[FaceSelectorGender] = list(get_args(FaceSelectorGender))
-face_parser_models : List[FaceParserModel] = [ 'bisenet_resnet_18', 'bisenet_resnet_34' ]
+face_selector_races : List[FaceSelectorRace] = list(get_args(FaceSelectorRace))
-face_mask_types : List[FaceMaskType] = [ 'box', 'occlusion', 'area', 'region' ]
+face_occluder_models : List[FaceOccluderModel] = list(get_args(FaceOccluderModel))
+face_parser_models : List[FaceParserModel] = list(get_args(FaceParserModel))
+face_mask_types : List[FaceMaskType] = list(get_args(FaceMaskType))
 face_mask_area_set : FaceMaskAreaSet =\
 {
 	'upper-face': [ 0, 1, 2, 31, 32, 33, 34, 35, 14, 15, 16, 26, 25, 24, 23, 22, 21, 20, 19, 18, 17 ],
@@ -40,10 +42,10 @@ face_mask_region_set : FaceMaskRegionSet =\
 	'upper-lip': 12,
 	'lower-lip': 13
 }
-face_mask_areas : List[FaceMaskArea] = list(face_mask_area_set.keys())
+face_mask_areas : List[FaceMaskArea] = list(get_args(FaceMaskArea))
-face_mask_regions : List[FaceMaskRegion] = list(face_mask_region_set.keys())
+face_mask_regions : List[FaceMaskRegion] = list(get_args(FaceMaskRegion))
 
-voice_extractor_models : List[VoiceExtractorModel] = [ 'kim_vocal_1', 'kim_vocal_2', 'uvr_mdxnet' ]
+voice_extractor_models : List[VoiceExtractorModel] = list(get_args(VoiceExtractorModel))
 
 audio_type_set : AudioTypeSet =\
 {
@@ -74,21 +76,21 @@ video_type_set : VideoTypeSet =\
 	'webm': 'video/webm',
 	'wmv': 'video/x-ms-wmv'
 }
-audio_formats : List[AudioFormat] = list(audio_type_set.keys())
+audio_formats : List[AudioFormat] = list(get_args(AudioFormat))
-image_formats : List[ImageFormat] = list(image_type_set.keys())
+image_formats : List[ImageFormat] = list(get_args(ImageFormat))
-video_formats : List[VideoFormat] = list(video_type_set.keys())
+video_formats : List[VideoFormat] = list(get_args(VideoFormat))
-temp_frame_formats : List[TempFrameFormat] = [ 'bmp', 'jpeg', 'png', 'tiff' ]
+temp_frame_formats : List[TempFrameFormat] = list(get_args(TempFrameFormat))
 
+output_audio_encoders : List[AudioEncoder] = list(get_args(AudioEncoder))
+output_video_encoders : List[VideoEncoder] = list(get_args(VideoEncoder))
 output_encoder_set : EncoderSet =\
 {
-	'audio': [ 'flac', 'aac', 'libmp3lame', 'libopus', 'libvorbis', 'pcm_s16le', 'pcm_s32le' ],
+	'audio': output_audio_encoders,
-	'video': [ 'libx264', 'libx264rgb', 'libx265', 'libvpx-vp9', 'h264_nvenc', 'hevc_nvenc', 'h264_amf', 'hevc_amf', 'h264_qsv', 'hevc_qsv', 'h264_videotoolbox', 'hevc_videotoolbox', 'rawvideo' ]
+	'video': output_video_encoders
 }
-output_audio_encoders : List[AudioEncoder] = output_encoder_set.get('audio')
+output_video_presets : List[VideoPreset] = list(get_args(VideoPreset))
-output_video_encoders : List[VideoEncoder] = output_encoder_set.get('video')
-output_video_presets : List[VideoPreset] = [ 'ultrafast', 'superfast', 'veryfast', 'faster', 'fast', 'medium', 'slow', 'slower', 'veryslow' ]
 
-benchmark_modes : List[BenchmarkMode] = [ 'warm', 'cold' ]
+benchmark_modes : List[BenchmarkMode] = list(get_args(BenchmarkMode))
 benchmark_set : BenchmarkSet =\
 {
 	'240p': '.assets/examples/target-240p.mp4',
@@ -99,20 +101,21 @@ benchmark_set : BenchmarkSet =\
 	'1440p': '.assets/examples/target-1440p.mp4',
 	'2160p': '.assets/examples/target-2160p.mp4'
 }
-benchmark_resolutions : List[BenchmarkResolution] = list(benchmark_set.keys())
+benchmark_resolutions : List[BenchmarkResolution] = list(get_args(BenchmarkResolution))
 
 execution_provider_set : ExecutionProviderSet =\
 {
 	'cuda': 'CUDAExecutionProvider',
 	'tensorrt': 'TensorrtExecutionProvider',
-	'directml': 'DmlExecutionProvider',
 	'rocm': 'ROCMExecutionProvider',
 	'migraphx': 'MIGraphXExecutionProvider',
-	'openvino': 'OpenVINOExecutionProvider',
 	'coreml': 'CoreMLExecutionProvider',
+	'openvino': 'OpenVINOExecutionProvider',
+	'qnn': 'QNNExecutionProvider',
+	'directml': 'DmlExecutionProvider',
 	'cpu': 'CPUExecutionProvider'
 }
-execution_providers : List[ExecutionProvider] = list(execution_provider_set.keys())
+execution_providers : List[ExecutionProvider] = list(get_args(ExecutionProvider))
 download_provider_set : DownloadProviderSet =\
 {
 	'github':
@@ -133,10 +136,10 @@ download_provider_set : DownloadProviderSet =\
 		'path': '/facefusion/{base_name}/resolve/main/{file_name}'
 	}
 }
-download_providers : List[DownloadProvider] = list(download_provider_set.keys())
+download_providers : List[DownloadProvider] = list(get_args(DownloadProvider))
-download_scopes : List[DownloadScope] = [ 'lite', 'full' ]
+download_scopes : List[DownloadScope] = list(get_args(DownloadScope))
 
-video_memory_strategies : List[VideoMemoryStrategy] = [ 'strict', 'moderate', 'tolerant' ]
+video_memory_strategies : List[VideoMemoryStrategy] = list(get_args(VideoMemoryStrategy))
 
 log_level_set : LogLevelSet =\
 {
@@ -145,14 +148,13 @@ log_level_set : LogLevelSet =\
 	'info': logging.INFO,
 	'debug': logging.DEBUG
 }
-log_levels : List[LogLevel] = list(log_level_set.keys())
+log_levels : List[LogLevel] = list(get_args(LogLevel))
 
-ui_workflows : List[UiWorkflow] = [ 'instant_runner', 'job_runner', 'job_manager' ]
+ui_workflows : List[UiWorkflow] = list(get_args(UiWorkflow))
-job_statuses : List[JobStatus] = [ 'drafted', 'queued', 'completed', 'failed' ]
+job_statuses : List[JobStatus] = list(get_args(JobStatus))
 
 benchmark_cycle_count_range : Sequence[int] = create_int_range(1, 10, 1)
 execution_thread_count_range : Sequence[int] = create_int_range(1, 32, 1)
-system_memory_limit_range : Sequence[int] = create_int_range(0, 128, 4)
 face_detector_margin_range : Sequence[int] = create_int_range(0, 100, 1)
 face_detector_angles : Sequence[Angle] = create_int_range(0, 270, 90)
 face_detector_score_range : Sequence[Score] = create_float_range(0.0, 1.0, 0.05)
@@ -161,6 +163,8 @@ face_mask_blur_range : Sequence[float] = create_float_range(0.0, 1.0, 0.05)
 face_mask_padding_range : Sequence[int] = create_int_range(0, 100, 1)
 face_selector_age_range : Sequence[int] = create_int_range(0, 100, 1)
 reference_face_distance_range : Sequence[float] = create_float_range(0.0, 1.0, 0.05)
+face_tracker_score_range : Sequence[Score] = create_float_range(0.0, 0.5, 0.05)
+target_frame_amount_range : Sequence[int] = create_int_range(0, 10, 1)
 output_image_quality_range : Sequence[int] = create_int_range(0, 100, 1)
 output_image_scale_range : Sequence[float] = create_float_range(0.25, 8.0, 0.25)
 output_audio_quality_range : Sequence[int] = create_int_range(0, 100, 1)

facefusion/common_helper.py

@@ -78,6 +78,12 @@ def get_first(__list__ : Any) -> Any:
 	return None
 
 
+def get_middle(__list__ : Any) -> Any:
+	if isinstance(__list__, Sequence) and __list__:
+		return __list__[len(__list__) // 2]
+	return None
+
+
 def get_last(__list__ : Any) -> Any:
 	if isinstance(__list__, Reversible):
 		return next(reversed(__list__), None)

facefusion/conda.py

@@ -0,0 +1,41 @@
+import os
+import sys
+from typing import List
+
+from facefusion.common_helper import is_linux, is_windows
+
+
+def setup() -> None:
+	conda_prefix = os.getenv('CONDA_PREFIX')
+	conda_ready = os.getenv('CONDA_READY')
+
+	if conda_prefix and not conda_ready:
+		if is_linux():
+			python_id = 'python' + str(sys.version_info.major) + '.' + str(sys.version_info.minor)
+			library_paths : List[str] =\
+			[
+				os.path.join(conda_prefix, 'lib'),
+				os.path.join(conda_prefix, 'lib', python_id, 'site-packages', 'tensorrt_libs')
+			]
+			library_paths = list(filter(os.path.exists, library_paths))
+
+			if library_paths:
+				if os.getenv('LD_LIBRARY_PATH'):
+					library_paths.append(os.getenv('LD_LIBRARY_PATH'))
+				os.environ['LD_LIBRARY_PATH'] = os.pathsep.join(library_paths)
+				os.environ['CONDA_READY'] = '1'
+				os.execv(sys.executable, [ sys.executable ] + sys.argv)
+
+		if is_windows():
+			library_paths =\
+			[
+				os.path.join(conda_prefix, 'Lib'),
+				os.path.join(conda_prefix, 'Lib', 'site-packages', 'tensorrt_libs')
+			]
+			library_paths = list(filter(os.path.exists, library_paths))
+
+			if library_paths:
+				if os.getenv('PATH'):
+					library_paths.append(os.getenv('PATH'))
+				os.environ['PATH'] = os.pathsep.join(library_paths)
+				os.environ['CONDA_READY'] = '1'

facefusion/config.py

@@ -1,29 +1,20 @@
 from configparser import ConfigParser
+from functools import lru_cache
 from typing import List, Optional
 
 from facefusion import state_manager
 from facefusion.common_helper import cast_bool, cast_float, cast_int
 
-CONFIG_PARSER = None
 
-
+@lru_cache
-def get_config_parser() -> ConfigParser:
+def get_static_config_parser() -> ConfigParser:
-	global CONFIG_PARSER
+	config_parser = ConfigParser()
-
+	config_parser.read(state_manager.get_item('config_path'), encoding = 'utf-8')
-	if CONFIG_PARSER is None:
+	return config_parser
-		CONFIG_PARSER = ConfigParser()
-		CONFIG_PARSER.read(state_manager.get_item('config_path'), encoding = 'utf-8')
-	return CONFIG_PARSER
-
-
-def clear_config_parser() -> None:
-	global CONFIG_PARSER
-
-	CONFIG_PARSER = None
 
 
 def get_str_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[str]:
-	config_parser = get_config_parser()
+	config_parser = get_static_config_parser()
 
 	if config_parser.has_option(section, option) and config_parser.get(section, option).strip():
 		return config_parser.get(section, option)
@@ -31,7 +22,7 @@ def get_str_value(section : str, option : str, fallback : Optional[str] = None)
 
 
 def get_int_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[int]:
-	config_parser = get_config_parser()
+	config_parser = get_static_config_parser()
 
 	if config_parser.has_option(section, option) and config_parser.get(section, option).strip():
 		return config_parser.getint(section, option)
@@ -39,7 +30,7 @@ def get_int_value(section : str, option : str, fallback : Optional[str] = None)
 
 
 def get_float_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[float]:
-	config_parser = get_config_parser()
+	config_parser = get_static_config_parser()
 
 	if config_parser.has_option(section, option) and config_parser.get(section, option).strip():
 		return config_parser.getfloat(section, option)
@@ -47,7 +38,7 @@ def get_float_value(section : str, option : str, fallback : Optional[str] = None
 
 
 def get_bool_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[bool]:
-	config_parser = get_config_parser()
+	config_parser = get_static_config_parser()
 
 	if config_parser.has_option(section, option) and config_parser.get(section, option).strip():
 		return config_parser.getboolean(section, option)
@@ -55,7 +46,7 @@ def get_bool_value(section : str, option : str, fallback : Optional[str] = None)
 
 
 def get_str_list(section : str, option : str, fallback : Optional[str] = None) -> Optional[List[str]]:
-	config_parser = get_config_parser()
+	config_parser = get_static_config_parser()
 
 	if config_parser.has_option(section, option) and config_parser.get(section, option).strip():
 		return config_parser.get(section, option).split()
@@ -65,7 +56,7 @@ def get_str_list(section : str, option : str, fallback : Optional[str] = None) -
 
 
 def get_int_list(section : str, option : str, fallback : Optional[str] = None) -> Optional[List[int]]:
-	config_parser = get_config_parser()
+	config_parser = get_static_config_parser()
 
 	if config_parser.has_option(section, option) and config_parser.get(section, option).strip():
 		return list(map(int, config_parser.get(section, option).split()))

facefusion/content_analyser.py

@@ -1,16 +1,14 @@
 from functools import lru_cache
-from typing import List, Tuple
+from typing import Tuple
 
 import numpy
 from tqdm import tqdm
 
 from facefusion import inference_manager, state_manager, translator
-from facefusion.common_helper import is_macos
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
-from facefusion.execution import has_execution_provider
 from facefusion.filesystem import resolve_relative_path
 from facefusion.thread_helper import conditional_thread_semaphore
-from facefusion.types import Detection, DownloadScope, DownloadSet, ExecutionProvider, Fps, InferencePool, ModelSet, VisionFrame
+from facefusion.types import Detection, DownloadScope, DownloadSet, Fps, InferencePool, ModelSet, VisionFrame
 from facefusion.vision import detect_video_fps, fit_contain_frame, read_image, read_video_frame
 
 STREAM_COUNTER = 0
@@ -119,12 +117,6 @@ def clear_inference_pool() -> None:
 	inference_manager.clear_inference_pool(__name__, model_names)
 
 
-def resolve_execution_providers() -> List[ExecutionProvider]:
-	if is_macos() and has_execution_provider('coreml'):
-		return [ 'cpu' ]
-	return state_manager.get_item('execution_providers')
-
-
 def collect_model_downloads() -> Tuple[DownloadSet, DownloadSet]:
 	model_set = create_static_model_set('full')
 	model_hash_set = {}
@@ -175,10 +167,12 @@ def analyse_video(video_path : str, trim_frame_start : int, trim_frame_end : int
 		for frame_number in frame_range:
 			if frame_number % int(video_fps) == 0:
 				vision_frame = read_video_frame(video_path, frame_number)
-				total += 1
 
-				if analyse_frame(vision_frame):
+				if numpy.any(vision_frame):
-					counter += 1
+					total += 1
+
+					if analyse_frame(vision_frame):
+						counter += 1
 
 			if counter > 0 and total > 0:
 				rate = counter / total * 100

facefusion/core.py

@@ -5,14 +5,13 @@ import signal
 import sys
 from time import time
 
-from facefusion import benchmarker, cli_helper, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, hash_helper, logger, state_manager, translator, voice_extractor
+from facefusion import benchmarker, cli_helper, content_analyser, hash_helper, logger, state_manager, translator
 from facefusion.args import apply_args, collect_job_args, reduce_job_args, reduce_step_args
 from facefusion.download import conditional_download_hashes, conditional_download_sources
 from facefusion.exit_helper import hard_exit, signal_exit
 from facefusion.filesystem import get_file_extension, get_file_name, is_image, is_video, resolve_file_paths, resolve_file_pattern
 from facefusion.jobs import job_helper, job_manager, job_runner
 from facefusion.jobs.job_list import compose_job_list
-from facefusion.memory import limit_system_memory
 from facefusion.processors.core import get_processors_modules
 from facefusion.program import create_program
 from facefusion.program_helper import validate_args
@@ -41,11 +40,6 @@ def cli() -> None:
 
 
 def route(args : Args) -> None:
-	system_memory_limit = state_manager.get_item('system_memory_limit')
-
-	if system_memory_limit and system_memory_limit > 0:
-		limit_system_memory(system_memory_limit)
-
 	if state_manager.get_item('command') == 'force-download':
 		error_code = force_download()
 		hard_exit(error_code)
@@ -107,21 +101,9 @@ def pre_check() -> bool:
 
 
 def common_pre_check() -> bool:
-	common_modules =\
-	[
-		content_analyser,
-		face_classifier,
-		face_detector,
-		face_landmarker,
-		face_masker,
-		face_recognizer,
-		voice_extractor
-	]
-
 	content_analyser_content = inspect.getsource(content_analyser).encode()
-	content_analyser_hash = hash_helper.create_hash(content_analyser_content)
 
-	return all(module.pre_check() for module in common_modules) and content_analyser_hash == 'b14e7b92'
+	return hash_helper.create_hash(content_analyser_content) == '975d67d6'
 
 
 def processors_pre_check() -> bool:
@@ -132,22 +114,19 @@ def processors_pre_check() -> bool:
 
 
 def force_download() -> ErrorCode:
-	common_modules =\
+	download_scope = state_manager.get_item('download_scope')
-	[
-		content_analyser,
-		face_classifier,
-		face_detector,
-		face_landmarker,
-		face_masker,
-		face_recognizer,
-		voice_extractor
-	]
 	available_processors = [ get_file_name(file_path) for file_path in resolve_file_paths('facefusion/processors/modules') ]
 	processor_modules = get_processors_modules(available_processors)
+	common_modules = []
+
+	for processor_module in processor_modules:
+		for common_module in processor_module.get_common_modules():
+			if common_module not in common_modules:
+				common_modules.append(common_module)
 
 	for module in common_modules + processor_modules:
 		if hasattr(module, 'create_static_model_set'):
-			for model in module.create_static_model_set(state_manager.get_item('download_scope')).values():
+			for model in module.create_static_model_set(download_scope).values():
 				model_hash_set = model.get('hashes')
 				model_source_set = model.get('sources')
 

facefusion/curl_builder.py

@@ -9,14 +9,14 @@ from facefusion.types import Command
 def run(commands : List[Command]) -> List[Command]:
 	user_agent = metadata.get('name') + '/' + metadata.get('version')
 
-	return [ shutil.which('curl'), '--user-agent', user_agent, '--insecure', '--location', '--silent' ] + commands
+	return [ shutil.which('curl'), '--user-agent', user_agent, '--location', '--silent', '--ssl-no-revoke' ] + commands
 
 
 def chain(*commands : List[Command]) -> List[Command]:
 	return list(itertools.chain(*commands))
 
 
-def head(url : str) -> List[Command]:
+def ping(url : str) -> List[Command]:
 	return [ '-I', url ]
 
 
@@ -26,3 +26,7 @@ def download(url : str, download_file_path : str) -> List[Command]:
 
 def set_timeout(timeout : int) -> List[Command]:
 	return [ '--connect-timeout', str(timeout) ]
+
+
+def set_retry(retry : int) -> List[Command]:
+	return [ '--retry', str(retry) ]

facefusion/download.py

@@ -29,7 +29,8 @@ def conditional_download(download_directory_path : str, urls : List[str]) -> Non
 			with tqdm(total = download_size, initial = initial_size, desc = translator.get('downloading'), unit = 'B', unit_scale = True, unit_divisor = 1024, ascii = ' =', disable = state_manager.get_item('log_level') in [ 'warn', 'error' ]) as progress:
 				commands = curl_builder.chain(
 					curl_builder.download(url, download_file_path),
-					curl_builder.set_timeout(5)
+					curl_builder.set_timeout(5),
+					curl_builder.set_retry(5)
 				)
 				open_curl(commands)
 				current_size = initial_size
@@ -44,7 +45,7 @@ def conditional_download(download_directory_path : str, urls : List[str]) -> Non
 @lru_cache(maxsize = 64)
 def get_static_download_size(url : str) -> int:
 	commands = curl_builder.chain(
-		curl_builder.head(url),
+		curl_builder.ping(url),
 		curl_builder.set_timeout(5)
 	)
 	process = open_curl(commands)
@@ -62,7 +63,7 @@ def get_static_download_size(url : str) -> int:
 @lru_cache(maxsize = 64)
 def ping_static_url(url : str) -> bool:
 	commands = curl_builder.chain(
-		curl_builder.head(url),
+		curl_builder.ping(url),
 		curl_builder.set_timeout(5)
 	)
 	process = open_curl(commands)

facefusion/execution.py

@@ -1,15 +1,17 @@
+import os
 import shutil
 import subprocess
 import xml.etree.ElementTree as ElementTree
 from functools import lru_cache
 from typing import List, Optional
 
-from onnxruntime import get_available_providers, set_default_logger_severity
+import onnxruntime
 
 import facefusion.choices
-from facefusion.types import ExecutionDevice, ExecutionProvider, InferenceSessionProvider, ValueAndUnit
+from facefusion.filesystem import create_directory, is_directory
+from facefusion.types import ExecutionDevice, ExecutionProvider, InferenceOptionSet, InferenceProvider, ValueAndUnit
 
-set_default_logger_severity(3)
+onnxruntime.set_default_logger_severity(3)
 
 
 def has_execution_provider(execution_provider : ExecutionProvider) -> bool:
@@ -17,7 +19,7 @@ def has_execution_provider(execution_provider : ExecutionProvider) -> bool:
 
 
 def get_available_execution_providers() -> List[ExecutionProvider]:
-	inference_session_providers = get_available_providers()
+	inference_session_providers = onnxruntime.get_available_providers()
 	available_execution_providers : List[ExecutionProvider] = []
 
 	for execution_provider, execution_provider_value in facefusion.choices.execution_provider_set.items():
@@ -28,54 +30,86 @@ def get_available_execution_providers() -> List[ExecutionProvider]:
 	return available_execution_providers
 
 
-def create_inference_session_providers(execution_device_id : int, execution_providers : List[ExecutionProvider]) -> List[InferenceSessionProvider]:
+def create_inference_providers(execution_device_id : int, execution_providers : List[ExecutionProvider]) -> List[InferenceProvider]:
-	inference_session_providers : List[InferenceSessionProvider] = []
+	inference_providers : List[InferenceProvider] = []
+	cache_path = resolve_cache_path()
 
 	for execution_provider in execution_providers:
 		if execution_provider == 'cuda':
-			inference_session_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
+			inference_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
 			{
 				'device_id': execution_device_id,
 				'cudnn_conv_algo_search': resolve_cudnn_conv_algo_search()
 			}))
+
 		if execution_provider == 'tensorrt':
-			inference_session_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
+			inference_option_set : InferenceOptionSet =\
 			{
-				'device_id': execution_device_id,
+				'device_id': execution_device_id
-				'trt_engine_cache_enable': True,
+			}
-				'trt_engine_cache_path': '.caches',
+			if is_directory(cache_path) or create_directory(cache_path):
-				'trt_timing_cache_enable': True,
+				inference_option_set.update(
-				'trt_timing_cache_path': '.caches',
+				{
-				'trt_builder_optimization_level': 5
+					'trt_engine_cache_enable': True,
-			}))
+					'trt_engine_cache_path': cache_path,
+					'trt_timing_cache_enable': True,
+					'trt_timing_cache_path': cache_path,
+					'trt_builder_optimization_level': 4
+				})
+			inference_providers.append((facefusion.choices.execution_provider_set.get(execution_provider), inference_option_set))
+
 		if execution_provider in [ 'directml', 'rocm' ]:
-			inference_session_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
+			inference_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
 			{
 				'device_id': execution_device_id
 			}))
+
 		if execution_provider == 'migraphx':
-			inference_session_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
+			inference_option_set =\
 			{
-				'device_id': execution_device_id,
+				'device_id': execution_device_id
-				'migraphx_model_cache_dir': '.caches'
+			}
-			}))
+			if is_directory(cache_path) or create_directory(cache_path):
+				inference_option_set.update(
+				{
+					'migraphx_model_cache_dir': cache_path
+				})
+			inference_providers.append((facefusion.choices.execution_provider_set.get(execution_provider), inference_option_set))
+
+		if execution_provider == 'coreml':
+			inference_option_set =\
+			{
+				'SpecializationStrategy': 'FastPrediction'
+			}
+			if is_directory(cache_path) or create_directory(cache_path):
+				inference_option_set.update(
+				{
+					'ModelCacheDirectory': cache_path
+				})
+			inference_providers.append((facefusion.choices.execution_provider_set.get(execution_provider), inference_option_set))
+
 		if execution_provider == 'openvino':
-			inference_session_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
+			inference_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
 			{
 				'device_type': resolve_openvino_device_type(execution_device_id),
 				'precision': 'FP32'
 			}))
-		if execution_provider == 'coreml':
+
-			inference_session_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
+		if execution_provider == 'qnn':
+			inference_providers.append((facefusion.choices.execution_provider_set.get(execution_provider),
 			{
-				'SpecializationStrategy': 'FastPrediction',
+				'device_id': execution_device_id,
-				'ModelCacheDirectory': '.caches'
+				'backend_type': 'htp'
 			}))
 
 	if 'cpu' in execution_providers:
-		inference_session_providers.append(facefusion.choices.execution_provider_set.get('cpu'))
+		inference_providers.append(facefusion.choices.execution_provider_set.get('cpu'))
 
-	return inference_session_providers
+	return inference_providers
+
+
+def resolve_cache_path() -> str:
+	return os.path.join('.caches', onnxruntime.get_version_string())
 
 
 def resolve_cudnn_conv_algo_search() -> str:

facefusion/face_classifier.py

@@ -20,7 +20,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 			'__metadata__':
 			{
 				'vendor': 'dchen236',
-				'license': 'Non-Commercial',
+				'license': 'CC-BY-4.0',
 				'year': 2021
 			},
 			'hashes':

facefusion/face_creator.py

@@ -2,14 +2,13 @@ from typing import List, Optional
 
 import numpy
 
-from facefusion import state_manager
+from facefusion import face_store, state_manager
-from facefusion.common_helper import get_first
+from facefusion.common_helper import get_first, get_middle
 from facefusion.face_classifier import classify_face
 from facefusion.face_detector import detect_faces, detect_faces_by_angle
-from facefusion.face_helper import apply_nms, convert_to_face_landmark_5, estimate_face_angle, get_nms_threshold
+from facefusion.face_helper import apply_nms, average_points, convert_to_face_landmark_5, estimate_face_angle, get_nms_threshold
 from facefusion.face_landmarker import detect_face_landmark, estimate_face_landmark_68_5
 from facefusion.face_recognizer import calculate_face_embedding
-from facefusion.face_store import get_static_faces, set_static_faces
 from facefusion.types import BoundingBox, Face, FaceLandmark5, FaceLandmarkSet, FaceScoreSet, Score, VisionFrame
 
 
@@ -47,7 +46,9 @@ def create_faces(vision_frame : VisionFrame, bounding_boxes : List[BoundingBox],
 		}
 		face_embedding, face_embedding_norm = calculate_face_embedding(vision_frame, face_landmark_set.get('5/68'))
 		gender, age, race = classify_face(vision_frame, face_landmark_set.get('5/68'))
+
 		faces.append(Face(
+			origin = 'detect',
 			bounding_box = bounding_box,
 			score_set = face_score_set,
 			landmark_set = face_landmark_set,
@@ -68,7 +69,102 @@ def get_one_face(faces : List[Face], position : int = 0) -> Optional[Face]:
 	return None
 
 
-def get_average_face(faces : List[Face]) -> Optional[Face]:
+def get_many_faces(vision_frames : List[VisionFrame]) -> List[Face]:
+	many_faces : List[Face] = []
+
+	for vision_frame in vision_frames:
+		if numpy.any(vision_frame):
+			all_bounding_boxes = []
+			all_face_scores = []
+			all_face_landmarks_5 = []
+
+			for face_detector_angle in state_manager.get_item('face_detector_angles'):
+				if face_detector_angle == 0:
+					bounding_boxes, face_scores, face_landmarks_5 = detect_faces(vision_frame)
+				else:
+					bounding_boxes, face_scores, face_landmarks_5 = detect_faces_by_angle(vision_frame, face_detector_angle)
+				all_bounding_boxes.extend(bounding_boxes)
+				all_face_scores.extend(face_scores)
+				all_face_landmarks_5.extend(face_landmarks_5)
+
+			if all_bounding_boxes and all_face_scores and all_face_landmarks_5 and state_manager.get_item('face_detector_score') > 0:
+				faces = create_faces(vision_frame, all_bounding_boxes, all_face_scores, all_face_landmarks_5)
+
+				if faces:
+					many_faces.extend(faces)
+
+	return many_faces
+
+
+def get_static_faces(vision_frames : List[VisionFrame]) -> List[Face]:
+	many_faces : List[Face] = []
+
+	for vision_frame in vision_frames:
+		faces = face_store.get_faces(vision_frame)
+
+		if not faces:
+			with face_store.resolve_lock(vision_frame):
+				faces = face_store.get_faces(vision_frame)
+
+				if not faces:
+					faces = get_many_faces([ vision_frame ])
+
+					if faces:
+						face_store.set_faces(vision_frame, faces)
+
+		many_faces.extend(faces)
+
+	return many_faces
+
+
+def refill_faces(faces : List[Optional[Face]]) -> List[Face]:
+	fill_faces = []
+	anchor_index_previous = -1
+
+	for index, face in enumerate(faces):
+		if face:
+			for gap_index in range(anchor_index_previous + 1, index):
+				average_factor = (gap_index - anchor_index_previous) / (index - anchor_index_previous)
+				average_face = average_face_geometry([faces[anchor_index_previous], face], average_factor)
+				fill_faces.append(average_face)
+
+			fill_faces.append(face)
+			anchor_index_previous = index
+
+	return fill_faces
+
+
+def average_face_geometry(faces : List[Face], average_factor : float) -> Face:
+	face_first = get_first(faces)
+	face_middle = get_middle(faces)
+	face_anchor = face_middle
+
+	if average_factor < 0.5:
+		face_anchor = face_first
+
+	landmark_set : FaceLandmarkSet =\
+	{
+		'5': average_points(face_first.landmark_set.get('5'), face_middle.landmark_set.get('5'), average_factor),
+		'5/68': average_points(face_first.landmark_set.get('5/68'), face_middle.landmark_set.get('5/68'), average_factor),
+		'68': average_points(face_first.landmark_set.get('68'), face_middle.landmark_set.get('68'), average_factor),
+		'68/5': average_points(face_first.landmark_set.get('68/5'), face_middle.landmark_set.get('68/5'), average_factor)
+	}
+
+	return Face(
+		origin = 'refill',
+		bounding_box = average_points(face_first.bounding_box, face_middle.bounding_box, average_factor),
+		score_set = face_anchor.score_set,
+		landmark_set = landmark_set,
+		angle = estimate_face_angle(landmark_set.get('68/5')),
+		embedding = face_anchor.embedding,
+		embedding_norm = face_anchor.embedding_norm,
+		gender = face_anchor.gender,
+		age = face_anchor.age,
+		race = face_anchor.race
+	)
+
+
+def average_face_identity(faces : List[Face]) -> Optional[Face]:
 	face_embeddings = []
 	face_embeddings_norm = []
 
@@ -80,6 +176,7 @@ def get_average_face(faces : List[Face]) -> Optional[Face]:
 			face_embeddings_norm.append(face.embedding_norm)
 
 		return Face(
+			origin = first_face.origin,
 			bounding_box = first_face.bounding_box,
 			score_set = first_face.score_set,
 			landmark_set = first_face.landmark_set,
@@ -93,37 +190,6 @@ def get_average_face(faces : List[Face]) -> Optional[Face]:
 	return None
 
 
-def get_many_faces(vision_frames : List[VisionFrame]) -> List[Face]:
-	many_faces : List[Face] = []
-
-	for vision_frame in vision_frames:
-		if numpy.any(vision_frame):
-			static_faces = get_static_faces(vision_frame)
-			if static_faces:
-				many_faces.extend(static_faces)
-			else:
-				all_bounding_boxes = []
-				all_face_scores = []
-				all_face_landmarks_5 = []
-
-				for face_detector_angle in state_manager.get_item('face_detector_angles'):
-					if face_detector_angle == 0:
-						bounding_boxes, face_scores, face_landmarks_5 = detect_faces(vision_frame)
-					else:
-						bounding_boxes, face_scores, face_landmarks_5 = detect_faces_by_angle(vision_frame, face_detector_angle)
-					all_bounding_boxes.extend(bounding_boxes)
-					all_face_scores.extend(face_scores)
-					all_face_landmarks_5.extend(face_landmarks_5)
-
-				if all_bounding_boxes and all_face_scores and all_face_landmarks_5 and state_manager.get_item('face_detector_score') > 0:
-					faces = create_faces(vision_frame, all_bounding_boxes, all_face_scores, all_face_landmarks_5)
-
-					if faces:
-						many_faces.extend(faces)
-						set_static_faces(vision_frame, faces)
-	return many_faces
-
-
 def scale_face(target_face : Face, target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> Face:
 	scale_x = temp_vision_frame.shape[1] / target_vision_frame.shape[1]
 	scale_y = temp_vision_frame.shape[0] / target_vision_frame.shape[0]

facefusion/face_detector.py

@@ -381,11 +381,11 @@ def detect_with_yunet(vision_frame : VisionFrame, face_detector_size : str) -> T
 			face_scores.extend(face_scores_raw[keep_indices])
 			face_landmarks_5_raw = numpy.concatenate(
 			[
-				face_landmarks_5_raw[:, [0, 1]] * feature_stride + anchors,
+				face_landmarks_5_raw[:, [ 0, 1 ]] * feature_stride + anchors,
-				face_landmarks_5_raw[:, [2, 3]] * feature_stride + anchors,
+				face_landmarks_5_raw[:, [ 2, 3 ]] * feature_stride + anchors,
-				face_landmarks_5_raw[:, [4, 5]] * feature_stride + anchors,
+				face_landmarks_5_raw[:, [ 4, 5 ]] * feature_stride + anchors,
-				face_landmarks_5_raw[:, [6, 7]] * feature_stride + anchors,
+				face_landmarks_5_raw[:, [ 6, 7 ]] * feature_stride + anchors,
-				face_landmarks_5_raw[:, [8, 9]] * feature_stride + anchors
+				face_landmarks_5_raw[:, [ 8, 9 ]] * feature_stride + anchors
 			], axis = -1).reshape(-1, 5, 2)
 
 			for face_landmark_raw_5 in face_landmarks_5_raw[keep_indices]:

facefusion/face_helper.py

@@ -81,7 +81,7 @@ def warp_face_by_face_landmark_5(temp_vision_frame : VisionFrame, face_landmark_
 
 
 def warp_face_by_bounding_box(temp_vision_frame : VisionFrame, bounding_box : BoundingBox, crop_size : Size) -> Tuple[VisionFrame, Matrix]:
-	source_points = numpy.array([ [ bounding_box[0], bounding_box[1] ], [bounding_box[2], bounding_box[1] ], [ bounding_box[0], bounding_box[3] ] ]).astype(numpy.float32)
+	source_points = numpy.array([ [ bounding_box[0], bounding_box[1] ], [ bounding_box[2], bounding_box[1] ], [ bounding_box[0], bounding_box[3] ] ]).astype(numpy.float32)
 	target_points = numpy.array([ [ 0, 0 ], [ crop_size[0], 0 ], [ 0, crop_size[1] ] ]).astype(numpy.float32)
 	affine_matrix = cv2.getAffineTransform(source_points, target_points)
 	if bounding_box[2] - bounding_box[0] > crop_size[0] or bounding_box[3] - bounding_box[1] > crop_size[1]:
@@ -247,10 +247,30 @@ def get_nms_threshold(face_detector_model : FaceDetectorModel, face_detector_ang
 
 
 def merge_matrix(temp_matrices : List[Matrix]) -> Matrix:
-	matrix = numpy.vstack([temp_matrices[0], [0, 0, 1]])
+	matrix = numpy.vstack([ temp_matrices[0], [ 0, 0, 1 ] ])
 
 	for temp_matrix in temp_matrices[1:]:
 		temp_matrix = numpy.vstack([ temp_matrix, [ 0, 0, 1 ] ])
 		matrix = numpy.dot(temp_matrix, matrix)
 
 	return matrix[:2, :]
+
+
+def calculate_bounding_box_overlap(bounding_box_a : BoundingBox, bounding_box_b : BoundingBox) -> float:
+	intersection_x1 = max(bounding_box_a[0], bounding_box_b[0])
+	intersection_y1 = max(bounding_box_a[1], bounding_box_b[1])
+	intersection_x2 = min(bounding_box_a[2], bounding_box_b[2])
+	intersection_y2 = min(bounding_box_a[3], bounding_box_b[3])
+	intersection = max(0, intersection_x2 - intersection_x1) * max(0, intersection_y2 - intersection_y1)
+	bounding_box_area = (bounding_box_a[2] - bounding_box_a[0]) * (bounding_box_a[3] - bounding_box_a[1])
+	reference_bounding_box_area = (bounding_box_b[2] - bounding_box_b[0]) * (bounding_box_b[3] - bounding_box_b[1])
+	union = bounding_box_area + reference_bounding_box_area - intersection
+
+	if union > 0:
+		return intersection / union
+
+	return 0.0
+
+
+def average_points(points_previous : Points, points_next : Points, average_factor : float) -> Points:
+	return points_previous * (1 - average_factor) + points_next * average_factor

facefusion/face_masker.py

@@ -233,7 +233,7 @@ def create_area_mask(crop_vision_frame : VisionFrame, face_landmark_68 : FaceLan
 
 	convex_hull = cv2.convexHull(face_landmark_68[landmark_points].astype(numpy.int32))
 	area_mask = numpy.zeros(crop_size).astype(numpy.float32)
-	cv2.fillConvexPoly(area_mask, convex_hull, 1.0) # type: ignore[call-overload]
+	cv2.fillConvexPoly(area_mask, convex_hull, 1.0) #type:ignore[call-overload]
 	area_mask = (cv2.GaussianBlur(area_mask.clip(0, 1), (0, 0), 5).clip(0.5, 1) - 0.5) * 2
 	return area_mask
 

facefusion/face_selector.py

@@ -2,26 +2,35 @@ from typing import List
 
 import numpy
 
+import facefusion.choices
 from facefusion import state_manager
-from facefusion.face_analyser import get_many_faces, get_one_face
+from facefusion.common_helper import get_first, get_middle
+from facefusion.face_creator import get_one_face, get_static_faces
+from facefusion.face_tracker import track_faces
 from facefusion.types import Face, FaceSelectorOrder, Gender, Race, Score, VisionFrame
 
 
-def select_faces(reference_vision_frame : VisionFrame, target_vision_frame : VisionFrame) -> List[Face]:
+def select_faces(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], target_vision_frames : List[VisionFrame]) -> List[Face]:
-	target_faces = get_many_faces([ target_vision_frame ])
+	source_faces = get_static_faces(source_vision_frames)
+
+	if state_manager.get_item('face_tracker_score') > 0:
+		target_faces = track_faces(target_vision_frames, state_manager.get_item('face_tracker_score'))
+	else:
+		target_faces = get_static_faces([ get_middle(target_vision_frames) ])
 
 	if state_manager.get_item('face_selector_mode') == 'many':
-		return sort_and_filter_faces(target_faces)
+		return sort_and_filter_faces(source_faces, target_faces)
 
 	if state_manager.get_item('face_selector_mode') == 'one':
-		target_face = get_one_face(sort_and_filter_faces(target_faces))
+		target_face = get_one_face(sort_and_filter_faces(source_faces, target_faces))
 		if target_face:
 			return [ target_face ]
 
 	if state_manager.get_item('face_selector_mode') == 'reference':
-		reference_faces = get_many_faces([ reference_vision_frame ])
+		reference_faces = get_static_faces([ reference_vision_frame ])
-		reference_faces = sort_and_filter_faces(reference_faces)
+		reference_faces = sort_and_filter_faces(source_faces, reference_faces)
 		reference_face = get_one_face(reference_faces, state_manager.get_item('reference_face_position'))
+
 		if reference_face:
 			match_faces = find_match_faces([ reference_face ], target_faces, state_manager.get_item('reference_face_distance'))
 			return match_faces
@@ -53,17 +62,33 @@ def calculate_face_distance(face : Face, reference_face : Face) -> float:
 	return 0
 
 
-def sort_and_filter_faces(faces : List[Face]) -> List[Face]:
+def sort_and_filter_faces(source_faces : List[Face], target_faces : List[Face]) -> List[Face]:
-	if faces:
+	if target_faces:
 		if state_manager.get_item('face_selector_order'):
-			faces = sort_faces_by_order(faces, state_manager.get_item('face_selector_order'))
+			target_faces = sort_faces_by_order(target_faces, state_manager.get_item('face_selector_order'))
-		if state_manager.get_item('face_selector_gender'):
+
-			faces = filter_faces_by_gender(faces, state_manager.get_item('face_selector_gender'))
+		face_selector_gender = state_manager.get_item('face_selector_gender')
-		if state_manager.get_item('face_selector_race'):
+		face_selector_race = state_manager.get_item('face_selector_race')
-			faces = filter_faces_by_race(faces, state_manager.get_item('face_selector_race'))
+
+		if source_faces and face_selector_gender == 'auto' or face_selector_race == 'auto':
+			source_face = get_first(sort_faces_by_order(source_faces, 'large-small'))
+
+			if source_face:
+				if face_selector_gender == 'auto':
+					face_selector_gender = source_face.gender
+				if face_selector_race == 'auto':
+					face_selector_race = source_face.race
+
+		if face_selector_gender in facefusion.choices.genders:
+			target_faces = filter_faces_by_gender(target_faces, face_selector_gender)
+
+		if face_selector_race in facefusion.choices.races:
+			target_faces = filter_faces_by_race(target_faces, face_selector_race)
+
 		if state_manager.get_item('face_selector_age_start') or state_manager.get_item('face_selector_age_end'):
-			faces = filter_faces_by_age(faces, state_manager.get_item('face_selector_age_start'), state_manager.get_item('face_selector_age_end'))
+			target_faces = filter_faces_by_age(target_faces, state_manager.get_item('face_selector_age_start'), state_manager.get_item('face_selector_age_end'))
-	return faces
+
+	return target_faces
 
 
 def sort_faces_by_order(faces : List[Face], order : FaceSelectorOrder) -> List[Face]:

facefusion/face_store.py

@@ -1,28 +1,42 @@
+import threading
 from typing import List, Optional
 
+import numpy
+
 from facefusion.hash_helper import create_hash
 from facefusion.types import Face, FaceStore, VisionFrame
 
-FACE_STORE : FaceStore =\
+FACE_STORE : FaceStore = {}
-{
-	'static_faces': {}
-}
 
 
-def get_face_store() -> FaceStore:
+def get_faces(vision_frame : VisionFrame) -> Optional[List[Face]]:
-	return FACE_STORE
+	if numpy.any(vision_frame):
+		vision_hash = create_hash(vision_frame.tobytes())
+
+		if FACE_STORE.get(vision_hash):
+			return FACE_STORE.get(vision_hash).get('faces')
+
+	return None
 
 
-def get_static_faces(vision_frame : VisionFrame) -> Optional[List[Face]]:
+def set_faces(vision_frame : VisionFrame, faces : List[Face]) -> None:
-	vision_hash = create_hash(vision_frame.tobytes())
+	if numpy.any(vision_frame):
-	return FACE_STORE.get('static_faces').get(vision_hash)
+		vision_hash = create_hash(vision_frame.tobytes())
+		FACE_STORE.setdefault(vision_hash,
+		{
+			'lock': threading.Lock()
+		})['faces'] = faces
 
 
-def set_static_faces(vision_frame : VisionFrame, faces : List[Face]) -> None:
+def resolve_lock(vision_frame : VisionFrame) -> threading.Lock:
-	vision_hash = create_hash(vision_frame.tobytes())
+	if numpy.any(vision_frame):
-	if vision_hash:
+		vision_hash = create_hash(vision_frame.tobytes())
-		FACE_STORE['static_faces'][vision_hash] = faces
+		return FACE_STORE.setdefault(vision_hash,
+		{
+			'lock': threading.Lock()
+		}).get('lock')
+	return threading.Lock()
 
 
-def clear_static_faces() -> None:
+def clear_faces() -> None:
-	FACE_STORE['static_faces'].clear()
+	FACE_STORE.clear()

facefusion/face_tracker.py

@@ -0,0 +1,61 @@
+from typing import List
+
+from facefusion.common_helper import get_first, get_last
+from facefusion.face_creator import get_static_faces, refill_faces
+from facefusion.face_helper import calculate_bounding_box_overlap
+from facefusion.types import Face, FaceTrack, Score, VisionFrame
+
+
+def track_faces(vision_frames : List[VisionFrame], score : Score) -> List[Face]:
+	target_index = len(vision_frames) // 2
+	face_tracks = create_face_tracks(vision_frames, score)
+	temp_faces = []
+
+	for face_track in face_tracks:
+		track_indices = sorted(face_track)
+		track_index_first = get_first(track_indices)
+		track_index_last = get_last(track_indices)
+		track_range = range(track_index_first, track_index_last + 1)
+
+		if target_index in track_range:
+			fill_faces = []
+
+			for index in track_range:
+				fill_faces.append(face_track.get(index))
+
+			temp_faces.append(refill_faces(fill_faces)[target_index - track_index_first])
+
+	return temp_faces
+
+
+def create_face_tracks(vision_frames : List[VisionFrame], score : Score) -> List[FaceTrack]:
+	face_tracks : List[FaceTrack] = []
+
+	for frame_index, vision_frame in enumerate(vision_frames):
+		for face in get_static_faces([ vision_frame ]):
+			face_track = select_face_track(face_tracks, face, score)
+
+			if face_track:
+				face_track[frame_index] = face
+			else:
+				face_tracks.append(
+				{
+					frame_index : face
+				})
+
+	return face_tracks
+
+
+def select_face_track(face_tracks : List[FaceTrack], face : Face, score : Score) -> FaceTrack:
+	select_track : FaceTrack = {}
+	select_score = score
+
+	for face_track in face_tracks:
+		track_face = face_track.get(get_last(face_track))
+		track_score = calculate_bounding_box_overlap(face.bounding_box, track_face.bounding_box)
+
+		if track_score > select_score:
+			select_score = track_score
+			select_track = face_track
+
+	return select_track

facefusion/ffmpeg.py

@@ -9,7 +9,7 @@ from tqdm import tqdm
 import facefusion.choices
 from facefusion import ffmpeg_builder, logger, process_manager, state_manager, translator
 from facefusion.filesystem import get_file_format, remove_file
-from facefusion.temp_helper import get_temp_file_path, get_temp_frames_pattern
+from facefusion.temp_helper import get_temp_file_path, get_temp_frame_pattern
 from facefusion.types import AudioBuffer, AudioEncoder, Command, EncoderSet, Fps, Resolution, UpdateProgress, VideoEncoder, VideoFormat
 from facefusion.vision import detect_video_duration, detect_video_fps, pack_resolution, predict_video_frame_total
 
@@ -109,14 +109,16 @@ def get_available_encoder_set() -> EncoderSet:
 
 def extract_frames(target_path : str, temp_video_resolution : Resolution, temp_video_fps : Fps, trim_frame_start : int, trim_frame_end : int) -> bool:
 	extract_frame_total = predict_video_frame_total(target_path, temp_video_fps, trim_frame_start, trim_frame_end)
-	temp_frames_pattern = get_temp_frames_pattern(target_path, '%08d')
+	temp_frame_pattern = get_temp_frame_pattern(target_path, '%08d')
 	commands = ffmpeg_builder.chain(
 		ffmpeg_builder.set_input(target_path),
 		ffmpeg_builder.set_media_resolution(pack_resolution(temp_video_resolution)),
 		ffmpeg_builder.set_frame_quality(0),
+		ffmpeg_builder.enforce_pixel_format('rgb24'),
 		ffmpeg_builder.select_frame_range(trim_frame_start, trim_frame_end, temp_video_fps),
 		ffmpeg_builder.prevent_frame_drop(),
-		ffmpeg_builder.set_output(temp_frames_pattern)
+		ffmpeg_builder.set_start_number(trim_frame_start),
+		ffmpeg_builder.set_output(temp_frame_pattern)
 	)
 
 	with tqdm(total = extract_frame_total, desc = translator.get('extracting'), unit = 'frame', ascii = ' =', disable = state_manager.get_item('log_level') in [ 'warn', 'error' ]) as progress:
@@ -172,6 +174,7 @@ def restore_audio(target_path : str, output_path : str, trim_frame_start : int,
 	temp_video_path = get_temp_file_path(target_path)
 	temp_video_format = cast(VideoFormat, get_file_format(temp_video_path))
 	temp_video_duration = detect_video_duration(temp_video_path)
+	output_video_format = cast(VideoFormat, get_file_format(output_path))
 
 	output_audio_encoder = fix_audio_encoder(temp_video_format, output_audio_encoder)
 	commands = ffmpeg_builder.chain(
@@ -185,6 +188,7 @@ def restore_audio(target_path : str, output_path : str, trim_frame_start : int,
 		ffmpeg_builder.select_media_stream('0:v:0'),
 		ffmpeg_builder.select_media_stream('1:a:0'),
 		ffmpeg_builder.set_video_duration(temp_video_duration),
+		ffmpeg_builder.set_faststart(output_video_format),
 		ffmpeg_builder.force_output(output_path)
 	)
 	return run_ffmpeg(commands).returncode == 0
@@ -197,6 +201,7 @@ def replace_audio(target_path : str, audio_path : str, output_path : str) -> boo
 	temp_video_path = get_temp_file_path(target_path)
 	temp_video_format = cast(VideoFormat, get_file_format(temp_video_path))
 	temp_video_duration = detect_video_duration(temp_video_path)
+	output_video_format = cast(VideoFormat, get_file_format(output_path))
 
 	output_audio_encoder = fix_audio_encoder(temp_video_format, output_audio_encoder)
 	commands = ffmpeg_builder.chain(
@@ -207,6 +212,7 @@ def replace_audio(target_path : str, audio_path : str, output_path : str) -> boo
 		ffmpeg_builder.set_audio_quality(output_audio_encoder, output_audio_quality),
 		ffmpeg_builder.set_audio_volume(output_audio_volume),
 		ffmpeg_builder.set_video_duration(temp_video_duration),
+		ffmpeg_builder.set_faststart(output_video_format),
 		ffmpeg_builder.force_output(output_path)
 	)
 	return run_ffmpeg(commands).returncode == 0
@@ -219,14 +225,16 @@ def merge_video(target_path : str, temp_video_fps : Fps, output_video_resolution
 	merge_frame_total = predict_video_frame_total(target_path, output_video_fps, trim_frame_start, trim_frame_end)
 	temp_video_path = get_temp_file_path(target_path)
 	temp_video_format = cast(VideoFormat, get_file_format(temp_video_path))
-	temp_frames_pattern = get_temp_frames_pattern(target_path, '%08d')
+	temp_frame_pattern = get_temp_frame_pattern(target_path, '%08d')
 
 	output_video_encoder = fix_video_encoder(temp_video_format, output_video_encoder)
 	commands = ffmpeg_builder.chain(
 		ffmpeg_builder.set_input_fps(temp_video_fps),
-		ffmpeg_builder.set_input(temp_frames_pattern),
+		ffmpeg_builder.set_start_number(trim_frame_start),
+		ffmpeg_builder.set_input(temp_frame_pattern),
 		ffmpeg_builder.set_media_resolution(pack_resolution(output_video_resolution)),
 		ffmpeg_builder.set_video_encoder(output_video_encoder),
+		ffmpeg_builder.set_video_tag(output_video_encoder, temp_video_format),
 		ffmpeg_builder.set_video_quality(output_video_encoder, output_video_quality),
 		ffmpeg_builder.set_video_preset(output_video_encoder, output_video_preset),
 		ffmpeg_builder.concat(
@@ -243,7 +251,8 @@ def merge_video(target_path : str, temp_video_fps : Fps, output_video_resolution
 
 
 def concat_video(output_path : str, temp_output_paths : List[str]) -> bool:
-	concat_video_path = tempfile.mktemp()
+	file_descriptor, concat_video_path = tempfile.mkstemp()
+	os.close(file_descriptor)
 
 	with open(concat_video_path, 'w') as concat_video_file:
 		for temp_output_path in temp_output_paths:
@@ -252,11 +261,13 @@ def concat_video(output_path : str, temp_output_paths : List[str]) -> bool:
 		concat_video_file.close()
 
 	output_path = os.path.abspath(output_path)
+	output_video_format = cast(VideoFormat, get_file_format(output_path))
 	commands = ffmpeg_builder.chain(
 		ffmpeg_builder.unsafe_concat(),
 		ffmpeg_builder.set_input(concat_video_file.name),
 		ffmpeg_builder.copy_video_encoder(),
 		ffmpeg_builder.copy_audio_encoder(),
+		ffmpeg_builder.set_faststart(output_video_format),
 		ffmpeg_builder.force_output(output_path)
 	)
 	process = run_ffmpeg(commands)

facefusion/ffmpeg_builder.py

@@ -5,7 +5,7 @@ from typing import List, Optional
 import numpy
 
 from facefusion.filesystem import get_file_format
-from facefusion.types import AudioEncoder, Command, CommandSet, Duration, Fps, StreamMode, VideoEncoder, VideoPreset
+from facefusion.types import AudioEncoder, Command, CommandSet, Duration, Fps, StreamMode, VideoEncoder, VideoFormat, VideoPreset
 
 
 def run(commands : List[Command]) -> List[Command]:
@@ -48,7 +48,11 @@ def set_input(input_path : str) -> List[Command]:
 
 
 def set_input_fps(input_fps : Fps) -> List[Command]:
-	return [ '-r', str(input_fps)]
+	return [ '-r', str(input_fps) ]
+
+
+def set_start_number(frame_number : int) -> List[Command]:
+	return [ '-start_number', str(frame_number) ]
 
 
 def set_output(output_path : str) -> List[Command]:
@@ -79,6 +83,10 @@ def unsafe_concat() -> List[Command]:
 	return [ '-f', 'concat', '-safe', '0' ]
 
 
+def enforce_pixel_format(pixel_format : str) -> List[Command]:
+	return [ '-pix_fmt', pixel_format ]
+
+
 def set_pixel_format(video_encoder : VideoEncoder) -> List[Command]:
 	if video_encoder == 'rawvideo':
 		return [ '-pix_fmt', 'rgb24' ]
@@ -183,6 +191,18 @@ def copy_video_encoder() -> List[Command]:
 	return set_video_encoder('copy')
 
 
+def set_faststart(video_format : VideoFormat) -> List[Command]:
+	if video_format in [ 'm4v', 'mov', 'mp4' ]:
+		return [ '-movflags', '+faststart' ]
+	return []
+
+
+def set_video_tag(video_encoder : VideoEncoder, video_format : VideoFormat) -> List[Command]:
+	if video_format in [ 'm4v', 'mov', 'mp4' ] and video_encoder in [ 'libx265', 'hevc_nvenc', 'hevc_amf', 'hevc_qsv', 'hevc_videotoolbox' ]:
+		return [ '-tag:v', 'hvc1' ]
+	return []
+
+
 def set_video_quality(video_encoder : VideoEncoder, video_quality : int) -> List[Command]:
 	if video_encoder in [ 'libx264', 'libx264rgb', 'libx265' ]:
 		video_compression = numpy.round(numpy.interp(video_quality, [ 0, 100 ], [ 51, 0 ])).astype(int).item()

facefusion/inference_manager.py

@@ -1,5 +1,6 @@
 import importlib
 import random
+from functools import lru_cache
 from time import sleep, time
 from typing import List
 
@@ -8,11 +9,11 @@ from onnxruntime import InferenceSession
 from facefusion import logger, process_manager, state_manager, translator
 from facefusion.app_context import detect_app_context
 from facefusion.common_helper import is_windows
-from facefusion.execution import create_inference_session_providers, has_execution_provider
+from facefusion.execution import create_inference_providers, has_execution_provider
 from facefusion.exit_helper import fatal_exit
 from facefusion.filesystem import get_file_name, is_file
 from facefusion.time_helper import calculate_end_time
-from facefusion.types import DownloadSet, ExecutionProvider, InferencePool, InferencePoolSet
+from facefusion.types import DownloadSet, ExecutionProvider, InferencePool, InferencePoolSet, InferenceProvider
 
 INFERENCE_POOL_SET : InferencePoolSet =\
 {
@@ -25,7 +26,7 @@ def get_inference_pool(module_name : str, model_names : List[str], model_source_
 	while process_manager.is_checking():
 		sleep(0.5)
 	execution_device_ids = state_manager.get_item('execution_device_ids')
-	execution_providers = resolve_execution_providers(module_name)
+	execution_providers = state_manager.get_item('execution_providers')
 	app_context = detect_app_context()
 
 	for execution_device_id in execution_device_ids:
@@ -36,26 +37,27 @@ def get_inference_pool(module_name : str, model_names : List[str], model_source_
 		if app_context == 'ui' and INFERENCE_POOL_SET.get('cli').get(inference_context):
 			INFERENCE_POOL_SET['ui'][inference_context] = INFERENCE_POOL_SET.get('cli').get(inference_context)
 		if not INFERENCE_POOL_SET.get(app_context).get(inference_context):
-			INFERENCE_POOL_SET[app_context][inference_context] = create_inference_pool(model_source_set, execution_device_id, execution_providers)
+			inference_providers = resolve_static_inference_providers(module_name, execution_device_id)
+			INFERENCE_POOL_SET[app_context][inference_context] = create_inference_pool(model_source_set, inference_providers)
 
 	current_inference_context = get_inference_context(module_name, model_names, random.choice(execution_device_ids), execution_providers)
 	return INFERENCE_POOL_SET.get(app_context).get(current_inference_context)
 
 
-def create_inference_pool(model_source_set : DownloadSet, execution_device_id : int, execution_providers : List[ExecutionProvider]) -> InferencePool:
+def create_inference_pool(model_source_set : DownloadSet, inference_providers : List[InferenceProvider]) -> InferencePool:
 	inference_pool : InferencePool = {}
 
 	for model_name in model_source_set.keys():
 		model_path = model_source_set.get(model_name).get('path')
 		if is_file(model_path):
-			inference_pool[model_name] = create_inference_session(model_path, execution_device_id, execution_providers)
+			inference_pool[model_name] = create_inference_session(model_path, inference_providers)
 
 	return inference_pool
 
 
 def clear_inference_pool(module_name : str, model_names : List[str]) -> None:
 	execution_device_ids = state_manager.get_item('execution_device_ids')
-	execution_providers = resolve_execution_providers(module_name)
+	execution_providers = state_manager.get_item('execution_providers')
 	app_context = detect_app_context()
 
 	if is_windows() and has_execution_provider('directml'):
@@ -67,13 +69,12 @@ def clear_inference_pool(module_name : str, model_names : List[str]) -> None:
 			del INFERENCE_POOL_SET[app_context][inference_context]
 
 
-def create_inference_session(model_path : str, execution_device_id : int, execution_providers : List[ExecutionProvider]) -> InferenceSession:
+def create_inference_session(model_path : str, inference_providers : List[InferenceProvider]) -> InferenceSession:
 	model_file_name = get_file_name(model_path)
 	start_time = time()
 
 	try:
-		inference_session_providers = create_inference_session_providers(execution_device_id, execution_providers)
+		inference_session = InferenceSession(model_path, providers = inference_providers)
-		inference_session = InferenceSession(model_path, providers = inference_session_providers)
 		logger.debug(translator.get('loading_model_succeeded').format(model_name = model_file_name, seconds = calculate_end_time(start_time)), __name__)
 		return inference_session
 
@@ -87,9 +88,15 @@ def get_inference_context(module_name : str, model_names : List[str], execution_
 	return inference_context
 
 
-def resolve_execution_providers(module_name : str) -> List[ExecutionProvider]:
+@lru_cache()
+def resolve_static_inference_providers(module_name : str, execution_device_id : int) -> List[InferenceProvider]:
 	module = importlib.import_module(module_name)
+	execution_providers = state_manager.get_item('execution_providers')
 
-	if hasattr(module, 'resolve_execution_providers'):
+	if hasattr(module, 'resolve_inference_providers'):
-		return getattr(module, 'resolve_execution_providers')()
+		inference_providers = getattr(module, 'resolve_inference_providers')()
-	return state_manager.get_item('execution_providers')
+
+		if inference_providers:
+			return inference_providers
+
+	return create_inference_providers(execution_device_id, execution_providers)

facefusion/installer.py

@@ -10,30 +10,34 @@ from types import FrameType
 from facefusion import metadata
 from facefusion.common_helper import is_linux, is_windows
 
-LOCALS =\
+LOCALES =\
 {
 	'install_dependency': 'install the {dependency} package',
+	'force_reinstall': 'force reinstall of packages',
 	'skip_conda': 'skip the conda environment check',
 	'conda_not_activated': 'conda is not activated'
 }
 ONNXRUNTIME_SET =\
 {
-	'default': ('onnxruntime', '1.23.2')
+	'default': ('onnxruntime', '1.26.0')
 }
 if is_windows() or is_linux():
-	ONNXRUNTIME_SET['cuda'] = ('onnxruntime-gpu', '1.23.2')
+	ONNXRUNTIME_SET['cuda'] = ('onnxruntime-gpu', '1.26.0')
-	ONNXRUNTIME_SET['openvino'] = ('onnxruntime-openvino', '1.23.0')
+	ONNXRUNTIME_SET['openvino'] = ('onnxruntime-openvino', '1.24.1')
 if is_windows():
-	ONNXRUNTIME_SET['directml'] = ('onnxruntime-directml', '1.23.0')
+	ONNXRUNTIME_SET['directml'] = ('onnxruntime-directml', '1.24.4')
+	ONNXRUNTIME_SET['qnn'] = ('onnxruntime-qnn', '1.24.4')
 if is_linux():
-	ONNXRUNTIME_SET['rocm'] = ('onnxruntime-rocm', '1.21.0')
+	ONNXRUNTIME_SET['migraphx'] = ('onnxruntime-migraphx', '1.25.0')
+	ONNXRUNTIME_SET['rocm'] = ('onnxruntime-rocm', '1.22.2.post1')
 
 
 def cli() -> None:
 	signal.signal(signal.SIGINT, signal_exit)
 	program = ArgumentParser(formatter_class = partial(HelpFormatter, max_help_position = 50))
-	program.add_argument('--onnxruntime', help = LOCALS.get('install_dependency').format(dependency = 'onnxruntime'), choices = ONNXRUNTIME_SET.keys(), required = True)
+	program.add_argument('onnxruntime', help = LOCALES.get('install_dependency').format(dependency = 'onnxruntime'), choices = ONNXRUNTIME_SET.keys())
-	program.add_argument('--skip-conda', help = LOCALS.get('skip_conda'), action = 'store_true')
+	program.add_argument('--force-reinstall', help = LOCALES.get('force_reinstall'), action = 'store_true')
+	program.add_argument('--skip-conda', help = LOCALES.get('skip_conda'), action = 'store_true')
 	program.add_argument('-v', '--version', version = metadata.get('name') + ' ' + metadata.get('version'), action = 'version')
 	run(program)
 
@@ -45,56 +49,26 @@ def signal_exit(signum : int, frame : FrameType) -> None:
 def run(program : ArgumentParser) -> None:
 	args = program.parse_args()
 	has_conda = 'CONDA_PREFIX' in os.environ
-	onnxruntime_name, onnxruntime_version = ONNXRUNTIME_SET.get(args.onnxruntime)
 
 	if not args.skip_conda and not has_conda:
-		sys.stdout.write(LOCALS.get('conda_not_activated') + os.linesep)
+		sys.stdout.write(LOCALES.get('conda_not_activated') + os.linesep)
 		sys.exit(1)
 
+	commands = [ shutil.which('pip'), 'install' ]
+
+	if args.force_reinstall:
+		commands.append('--force-reinstall')
+
 	with open('requirements.txt') as file:
 
 		for line in file.readlines():
 			__line__ = line.strip()
 			if not __line__.startswith('onnxruntime'):
-				subprocess.call([ shutil.which('pip'), 'install', line, '--force-reinstall' ])
+				commands.append(__line__)
 
-	if args.onnxruntime == 'rocm':
+	onnxruntime_name, onnxruntime_version = ONNXRUNTIME_SET.get(args.onnxruntime)
-		python_id = 'cp' + str(sys.version_info.major) + str(sys.version_info.minor)
+	commands.append(onnxruntime_name + '==' + onnxruntime_version)
 
-		if python_id in [ 'cp310', 'cp312' ]:
+	subprocess.call([ shutil.which('pip'), 'uninstall', 'onnxruntime', onnxruntime_name, '-y', '-q' ])
-			wheel_name = 'onnxruntime_rocm-' + onnxruntime_version + '-' + python_id + '-' + python_id + '-linux_x86_64.whl'
-			wheel_url = 'https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4/' + wheel_name
-			subprocess.call([ shutil.which('pip'), 'install', wheel_url, '--force-reinstall' ])
-	else:
-		subprocess.call([ shutil.which('pip'), 'install', onnxruntime_name + '==' + onnxruntime_version, '--force-reinstall' ])
-
-	if args.onnxruntime == 'cuda' and has_conda:
-		library_paths = []
-
-		if is_linux():
-			if os.getenv('LD_LIBRARY_PATH'):
-				library_paths = os.getenv('LD_LIBRARY_PATH').split(os.pathsep)
-
-			python_id = 'python' + str(sys.version_info.major) + '.' + str(sys.version_info.minor)
-			library_paths.extend(
-			[
-				os.path.join(os.getenv('CONDA_PREFIX'), 'lib'),
-				os.path.join(os.getenv('CONDA_PREFIX'), 'lib', python_id, 'site-packages', 'tensorrt_libs')
-			])
-			library_paths = list(dict.fromkeys([ library_path for library_path in library_paths if os.path.exists(library_path) ]))
-
-			subprocess.call([ shutil.which('conda'), 'env', 'config', 'vars', 'set', 'LD_LIBRARY_PATH=' + os.pathsep.join(library_paths) ])
-
-		if is_windows():
-			if os.getenv('PATH'):
-				library_paths = os.getenv('PATH').split(os.pathsep)
-
-			library_paths.extend(
-			[
-				os.path.join(os.getenv('CONDA_PREFIX'), 'Lib'),
-				os.path.join(os.getenv('CONDA_PREFIX'), 'Lib', 'site-packages', 'tensorrt_libs')
-			])
-			library_paths = list(dict.fromkeys([ library_path for library_path in library_paths if os.path.exists(library_path) ]))
-
-			subprocess.call([ shutil.which('conda'), 'env', 'config', 'vars', 'set', 'PATH=' + os.pathsep.join(library_paths) ])
 
+	subprocess.call(commands)

facefusion/jobs/job_manager.py

@@ -6,6 +6,7 @@ import facefusion.choices
 from facefusion.filesystem import create_directory, get_file_name, is_directory, is_file, move_file, remove_directory, remove_file, resolve_file_pattern
 from facefusion.jobs.job_helper import get_step_output_path
 from facefusion.json import read_json, write_json
+from facefusion.sanitizer import sanitize_job_id
 from facefusion.time_helper import get_current_date_time
 from facefusion.types import Args, Job, JobSet, JobStatus, JobStep, JobStepStatus
 
@@ -261,5 +262,6 @@ def find_job_path(job_id : str) -> Optional[str]:
 
 def get_job_file_name(job_id : str) -> Optional[str]:
 	if job_id:
+		job_id = sanitize_job_id(job_id)
 		return job_id + '.json'
 	return None

facefusion/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{
@@ -124,6 +124,7 @@ LOCALS : Locals =\
 			'reference_face_position': 'specify the position used to create the reference face',
 			'reference_face_distance': 'specify the similarity between the reference face and target face',
 			'reference_frame_number': 'specify the frame used to create the reference face',
+			'face_tracker_score': 'specify the overlap score used to match the tracked faces',
 			'face_occluder_model': 'choose the model responsible for the occlusion mask',
 			'face_parser_model': 'choose the model responsible for the region mask',
 			'face_mask_types': 'mix and match different face mask types (choices: {choices})',
@@ -136,6 +137,7 @@ LOCALS : Locals =\
 			'trim_frame_end': 'specify the ending frame of the target video',
 			'temp_frame_format': 'specify the temporary resources format',
 			'keep_temp': 'keep the temporary resources after processing',
+			'target_frame_amount': 'specify the amount of target frames forwarded to the processor',
 			'output_image_quality': 'specify the image quality which translates to the image compression',
 			'output_image_scale': 'specify the image scale based on the target image',
 			'output_audio_encoder': 'specify the encoder used for the audio',
@@ -161,7 +163,6 @@ LOCALS : Locals =\
 			'execution_providers': 'inference using different providers (choices: {choices}, ...)',
 			'execution_thread_count': 'specify the amount of parallel threads while processing',
 			'video_memory_strategy': 'balance fast processing and low VRAM usage',
-			'system_memory_limit': 'limit the available RAM that can be used while processing',
 			'log_level': 'adjust the message severity displayed in the terminal',
 			'halt_on_error': 'halt the program once an error occurred',
 			'run': 'run the program',
@@ -189,7 +190,7 @@ LOCALS : Locals =\
 		},
 		'about':
 		{
-			'fund': 'fund training server',
+			'fund': 'fund ai workstation',
 			'subscribe': 'become a member',
 			'join': 'join our community'
 		},
@@ -224,6 +225,7 @@ LOCALS : Locals =\
 			'face_selector_mode_dropdown': 'FACE SELECTOR MODE',
 			'face_selector_order_dropdown': 'FACE SELECTOR ORDER',
 			'face_selector_race_dropdown': 'FACE SELECTOR RACE',
+			'face_tracker_score_slider': 'FACE TRACKER SCORE',
 			'face_occluder_model_dropdown': 'FACE OCCLUDER MODEL',
 			'face_parser_model_dropdown': 'FACE PARSER MODEL',
 			'voice_extractor_model_dropdown': 'VOICE EXTRACTOR MODEL',
@@ -257,7 +259,6 @@ LOCALS : Locals =\
 			'source_file': 'SOURCE',
 			'start_button': 'START',
 			'stop_button': 'STOP',
-			'system_memory_limit_slider': 'SYSTEM MEMORY LIMIT',
 			'target_file': 'TARGET',
 			'temp_frame_format_dropdown': 'TEMP FRAME FORMAT',
 			'terminal_textbox': 'TERMINAL',

facefusion/memory.py

@@ -1,21 +0,0 @@
-from facefusion.common_helper import is_macos, is_windows
-
-if is_windows():
-	import ctypes
-else:
-	import resource
-
-
-def limit_system_memory(system_memory_limit : int = 1) -> bool:
-	if is_macos():
-		system_memory_limit = system_memory_limit * (1024 ** 6)
-	else:
-		system_memory_limit = system_memory_limit * (1024 ** 3)
-	try:
-		if is_windows():
-			ctypes.windll.kernel32.SetProcessWorkingSetSize(-1, ctypes.c_size_t(system_memory_limit), ctypes.c_size_t(system_memory_limit)) #type:ignore[attr-defined]
-		else:
-			resource.setrlimit(resource.RLIMIT_DATA, (system_memory_limit, system_memory_limit))
-		return True
-	except Exception:
-		return False

facefusion/metadata.py

@@ -4,7 +4,7 @@ METADATA =\
 {
 	'name': 'FaceFusion',
 	'description': 'Industry leading face manipulation platform',
-	'version': '3.5.1',
+	'version': '3.7.1',
 	'license': 'OpenRAIL-AS',
 	'author': 'Henry Ruhs',
 	'url': 'https://facefusion.io'

facefusion/normalizer.py

@@ -19,9 +19,9 @@ def normalize_space(spaces : Optional[List[int]]) -> Optional[Padding]:
 	if spaces and len(spaces) == 1:
 		return tuple([spaces[0]] * 4) #type:ignore[return-value]
 	if spaces and len(spaces) == 2:
-		return tuple([spaces[0], spaces[1], spaces[0], spaces[1]]) #type:ignore[return-value]
+		return tuple([ spaces[0], spaces[1], spaces[0], spaces[1] ]) #type:ignore[return-value]
 	if spaces and len(spaces) == 3:
-		return tuple([spaces[0], spaces[1], spaces[2], spaces[1]]) #type:ignore[return-value]
+		return tuple([ spaces[0], spaces[1], spaces[2], spaces[1] ]) #type:ignore[return-value]
 	if spaces and len(spaces) == 4:
 		return tuple(spaces) #type:ignore[return-value]
 	return None

facefusion/processors/choices.py

@@ -1,27 +0,0 @@
-from facefusion.processors.modules.age_modifier.choices import age_modifier_direction_range, age_modifier_models  # noqa: F401
-from facefusion.processors.modules.background_remover.choices import background_remover_color_range, background_remover_models  # noqa: F401
-from facefusion.processors.modules.deep_swapper.choices import deep_swapper_models, deep_swapper_morph_range  # noqa: F401
-from facefusion.processors.modules.expression_restorer.choices import expression_restorer_areas, expression_restorer_factor_range, expression_restorer_models  # noqa: F401
-from facefusion.processors.modules.face_debugger.choices import face_debugger_items  # noqa: F401
-from facefusion.processors.modules.face_editor.choices import (  # noqa: F401
-	face_editor_eye_gaze_horizontal_range,
-	face_editor_eye_gaze_vertical_range,
-	face_editor_eye_open_ratio_range,
-	face_editor_eyebrow_direction_range,
-	face_editor_head_pitch_range,
-	face_editor_head_roll_range,
-	face_editor_head_yaw_range,
-	face_editor_lip_open_ratio_range,
-	face_editor_models,
-	face_editor_mouth_grim_range,
-	face_editor_mouth_position_horizontal_range,
-	face_editor_mouth_position_vertical_range,
-	face_editor_mouth_pout_range,
-	face_editor_mouth_purse_range,
-	face_editor_mouth_smile_range,
-)
-from facefusion.processors.modules.face_enhancer.choices import face_enhancer_blend_range, face_enhancer_models, face_enhancer_weight_range  # noqa: F401
-from facefusion.processors.modules.face_swapper.choices import face_swapper_models, face_swapper_set, face_swapper_weight_range  # noqa: F401
-from facefusion.processors.modules.frame_colorizer.choices import frame_colorizer_blend_range, frame_colorizer_models, frame_colorizer_sizes  # noqa: F401
-from facefusion.processors.modules.frame_enhancer.choices import frame_enhancer_blend_range, frame_enhancer_models  # noqa: F401
-from facefusion.processors.modules.lip_syncer.choices import lip_syncer_models, lip_syncer_weight_range  # noqa: F401

facefusion/processors/core.py

@@ -12,6 +12,7 @@ PROCESSORS_METHODS =\
 	'clear_inference_pool',
 	'register_args',
 	'apply_args',
+	'get_common_modules',
 	'pre_check',
 	'pre_process',
 	'post_process',

facefusion/processors/modules/age_modifier/choices.py

@@ -1,8 +1,8 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_int_range
 from facefusion.processors.modules.age_modifier.types import AgeModifierModel
 
-age_modifier_models : List[AgeModifierModel] = [ 'styleganex_age' ]
+age_modifier_models : List[AgeModifierModel] = list(get_args(AgeModifierModel))
 
 age_modifier_direction_range : Sequence[int] = create_int_range(-100, 100, 1)

facefusion/processors/modules/age_modifier/core.py

@@ -1,5 +1,7 @@
 from argparse import ArgumentParser
 from functools import lru_cache
+from types import ModuleType
+from typing import List
 
 import cv2
 import numpy
@@ -8,10 +10,9 @@ import facefusion.choices
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager
-from facefusion.common_helper import create_int_metavar, is_macos
+from facefusion.common_helper import create_int_metavar, get_middle
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
-from facefusion.execution import has_execution_provider
+from facefusion.face_creator import scale_face
-from facefusion.face_analyser import scale_face
 from facefusion.face_helper import merge_matrix, paste_back, scale_face_landmark_5, warp_face_by_face_landmark_5
 from facefusion.face_masker import create_box_mask, create_occlusion_mask
 from facefusion.face_selector import select_faces
@@ -22,13 +23,48 @@ from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import thread_semaphore
 from facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import match_frame_color, read_static_image, read_static_video_frame
+from facefusion.vision import match_frame_color, read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
 def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 	return\
 	{
+		'fran':
+		{
+			'__metadata__':
+			{
+				'vendor': 'ry-lu',
+				'license': 'mit',
+				'year': 2024
+			},
+			'hashes':
+			{
+				'age_modifier':
+				{
+					'url': resolve_download_url('models-3.6.0', 'fran.hash'),
+					'path': resolve_relative_path('../.assets/models/fran.hash')
+				}
+			},
+			'sources':
+			{
+				'age_modifier':
+				{
+					'url': resolve_download_url('models-3.6.0', 'fran.onnx'),
+					'path': resolve_relative_path('../.assets/models/fran.onnx')
+				}
+			},
+			'templates':
+			{
+				'target': 'ffhq_512',
+			},
+			'sizes':
+			{
+				'target': (1024, 1024),
+			},
+			'mean': [ 0.0, 0.0, 0.0 ],
+			'standard_deviation': [ 1.0, 1.0, 1.0 ]
+		},
 		'styleganex_age':
 		{
 			'__metadata__':
@@ -62,7 +98,9 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 			{
 				'target': (256, 256),
 				'target_with_background': (384, 384)
-			}
+			},
+			'mean': [ 0.5, 0.5, 0.5 ],
+			'standard_deviation': [ 0.5, 0.5, 0.5 ]
 		}
 	}
 
@@ -87,7 +125,7 @@ def get_model_options() -> ModelOptions:
 def register_args(program : ArgumentParser) -> None:
 	group_processors = find_argument_group(program, 'processors')
 	if group_processors:
-		group_processors.add_argument('--age-modifier-model', help = translator.get('help.model', __package__), default = config.get_str_value('processors', 'age_modifier_model', 'styleganex_age'), choices = age_modifier_choices.age_modifier_models)
+		group_processors.add_argument('--age-modifier-model', help = translator.get('help.model', __package__), default = config.get_str_value('processors', 'age_modifier_model', 'fran'), choices = age_modifier_choices.age_modifier_models)
 		group_processors.add_argument('--age-modifier-direction', help = translator.get('help.direction', __package__), type = int, default = config.get_int_value('processors', 'age_modifier_direction', '0'), choices = age_modifier_choices.age_modifier_direction_range, metavar = create_int_metavar(age_modifier_choices.age_modifier_direction_range))
 		facefusion.jobs.job_store.register_step_keys([ 'age_modifier_model', 'age_modifier_direction' ])
 
@@ -97,10 +135,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('age_modifier_direction', args.get('age_modifier_direction'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -120,16 +166,15 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
 
 
 def modify_age(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
@@ -137,41 +182,63 @@ def modify_age(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFra
 	model_sizes = get_model_options().get('sizes')
 	face_landmark_5 = target_face.landmark_set.get('5/68').copy()
 	crop_vision_frame, affine_matrix = warp_face_by_face_landmark_5(temp_vision_frame, face_landmark_5, model_templates.get('target'), model_sizes.get('target'))
-	extend_face_landmark_5 = scale_face_landmark_5(face_landmark_5, 0.875)
-	extend_vision_frame, extend_affine_matrix = warp_face_by_face_landmark_5(temp_vision_frame, extend_face_landmark_5, model_templates.get('target_with_background'), model_sizes.get('target_with_background'))
-	extend_vision_frame_raw = extend_vision_frame.copy()
-	box_mask = create_box_mask(extend_vision_frame, state_manager.get_item('face_mask_blur'), (0, 0, 0, 0))
-	crop_masks =\
-	[
-		box_mask
-	]
 
-	if 'occlusion' in state_manager.get_item('face_mask_types'):
+	if state_manager.get_item('age_modifier_model') == 'fran':
-		occlusion_mask = create_occlusion_mask(crop_vision_frame)
+		box_mask = create_box_mask(crop_vision_frame, state_manager.get_item('face_mask_blur'), (0, 0, 0, 0))
-		temp_matrix = merge_matrix([ extend_affine_matrix, cv2.invertAffineTransform(affine_matrix) ])
+		crop_masks =\
-		occlusion_mask = cv2.warpAffine(occlusion_mask, temp_matrix, model_sizes.get('target_with_background'))
+		[
-		crop_masks.append(occlusion_mask)
+			box_mask
+		]
 
-	crop_vision_frame = prepare_vision_frame(crop_vision_frame)
+		if 'occlusion' in state_manager.get_item('face_mask_types'):
-	extend_vision_frame = prepare_vision_frame(extend_vision_frame)
+			occlusion_mask = create_occlusion_mask(crop_vision_frame)
-	age_modifier_direction = numpy.array(numpy.interp(state_manager.get_item('age_modifier_direction'), [ -100, 100 ], [ 2.5, -2.5 ])).astype(numpy.float32)
+			crop_masks.append(occlusion_mask)
-	extend_vision_frame = forward(crop_vision_frame, extend_vision_frame, age_modifier_direction)
+
-	extend_vision_frame = normalize_extend_frame(extend_vision_frame)
+		crop_vision_frame = prepare_vision_frame(crop_vision_frame)
-	extend_vision_frame = match_frame_color(extend_vision_frame_raw, extend_vision_frame)
+		target_age = numpy.mean(target_face.age)
-	extend_affine_matrix *= (model_sizes.get('target')[0] * 4) / model_sizes.get('target_with_background')[0]
+		age_modifier_direction = numpy.array([ target_age, target_age + state_manager.get_item('age_modifier_direction') ], dtype = numpy.float32) / 100
-	crop_mask = numpy.minimum.reduce(crop_masks).clip(0, 1)
+		age_modifier_direction = age_modifier_direction.clip(0, 1)
-	crop_mask = cv2.resize(crop_mask, (model_sizes.get('target')[0] * 4, model_sizes.get('target')[1] * 4))
+		crop_vision_frame = forward(crop_vision_frame, crop_vision_frame, age_modifier_direction)
-	paste_vision_frame = paste_back(temp_vision_frame, extend_vision_frame, crop_mask, extend_affine_matrix)
+		crop_vision_frame = normalize_vision_frame(crop_vision_frame)
-	return paste_vision_frame
+		crop_mask = numpy.minimum.reduce(crop_masks).clip(0, 1)
+		paste_vision_frame = paste_back(temp_vision_frame, crop_vision_frame, crop_mask, affine_matrix)
+		return paste_vision_frame
+
+	if state_manager.get_item('age_modifier_model') == 'styleganex_age':
+		extend_face_landmark_5 = scale_face_landmark_5(face_landmark_5, 0.875)
+		extend_vision_frame, extend_affine_matrix = warp_face_by_face_landmark_5(temp_vision_frame, extend_face_landmark_5, model_templates.get('target_with_background'), model_sizes.get('target_with_background'))
+		extend_vision_frame_raw = extend_vision_frame.copy()
+		box_mask = create_box_mask(extend_vision_frame, state_manager.get_item('face_mask_blur'), (0, 0, 0, 0))
+		crop_masks =\
+		[
+			box_mask
+		]
+
+		if 'occlusion' in state_manager.get_item('face_mask_types'):
+			occlusion_mask = create_occlusion_mask(crop_vision_frame)
+			temp_matrix = merge_matrix([ extend_affine_matrix, cv2.invertAffineTransform(affine_matrix) ])
+			occlusion_mask = cv2.warpAffine(occlusion_mask, temp_matrix, model_sizes.get('target_with_background'))
+			crop_masks.append(occlusion_mask)
+
+		crop_vision_frame = prepare_vision_frame(crop_vision_frame)
+		extend_vision_frame = prepare_vision_frame(extend_vision_frame)
+		age_modifier_direction = numpy.array(numpy.interp(state_manager.get_item('age_modifier_direction'), [ -100, 100 ], [ 2.5, -2.5 ])).astype(numpy.float32)
+		extend_vision_frame = forward(crop_vision_frame, extend_vision_frame, age_modifier_direction)
+		extend_vision_frame = normalize_extend_frame(extend_vision_frame)
+		extend_vision_frame = match_frame_color(extend_vision_frame_raw, extend_vision_frame)
+		extend_affine_matrix *= (model_sizes.get('target')[0] * 4) / model_sizes.get('target_with_background')[0]
+		crop_mask = numpy.minimum.reduce(crop_masks).clip(0, 1)
+		crop_mask = cv2.resize(crop_mask, (model_sizes.get('target')[0] * 4, model_sizes.get('target')[1] * 4))
+		paste_vision_frame = paste_back(temp_vision_frame, extend_vision_frame, crop_mask, extend_affine_matrix)
+		return paste_vision_frame
+
+	return temp_vision_frame
 
 
 def forward(crop_vision_frame : VisionFrame, extend_vision_frame : VisionFrame, age_modifier_direction : AgeModifierDirection) -> VisionFrame:
 	age_modifier = get_inference_pool().get('age_modifier')
 	age_modifier_inputs = {}
 
-	if is_macos() and has_execution_provider('coreml'):
-		age_modifier.set_providers([ facefusion.choices.execution_provider_set.get('cpu') ])
-
 	for age_modifier_input in age_modifier.get_inputs():
 		if age_modifier_input.name == 'target':
 			age_modifier_inputs[age_modifier_input.name] = crop_vision_frame
@@ -187,12 +254,24 @@ def forward(crop_vision_frame : VisionFrame, extend_vision_frame : VisionFrame,
 
 
 def prepare_vision_frame(vision_frame : VisionFrame) -> VisionFrame:
+	model_mean = get_model_options().get('mean')
+	model_standard_deviation = get_model_options().get('standard_deviation')
 	vision_frame = vision_frame[:, :, ::-1] / 255.0
-	vision_frame = (vision_frame - 0.5) / 0.5
+	vision_frame = (vision_frame - model_mean) / model_standard_deviation
 	vision_frame = numpy.expand_dims(vision_frame.transpose(2, 0, 1), axis = 0).astype(numpy.float32)
 	return vision_frame
 
 
+def normalize_vision_frame(vision_frame : VisionFrame) -> VisionFrame:
+	model_mean = get_model_options().get('mean')
+	model_standard_deviation = get_model_options().get('standard_deviation')
+	vision_frame = vision_frame.transpose(1, 2, 0)
+	vision_frame = vision_frame * model_standard_deviation + model_mean
+	vision_frame = vision_frame.clip(0, 1)
+	vision_frame = vision_frame[:, :, ::-1] * 255
+	return vision_frame
+
+
 def normalize_extend_frame(extend_vision_frame : VisionFrame) -> VisionFrame:
 	model_sizes = get_model_options().get('sizes')
 	extend_vision_frame = numpy.clip(extend_vision_frame, -1, 1)
@@ -206,10 +285,13 @@ def normalize_extend_frame(extend_vision_frame : VisionFrame) -> VisionFrame:
 
 def process_frame(inputs : AgeModifierInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
-	target_vision_frame = inputs.get('target_vision_frame')
+	source_vision_frames = inputs.get('source_vision_frames')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+
+	target_vision_frame = get_middle(target_vision_frames)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if target_faces:
 		for target_face in target_faces:

facefusion/processors/modules/age_modifier/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/age_modifier/types.py

@@ -1,4 +1,4 @@
-from typing import Any, Literal, TypeAlias, TypedDict
+from typing import Any, List, Literal, TypeAlias, TypedDict
 
 from numpy.typing import NDArray
 
@@ -7,11 +7,12 @@ from facefusion.types import Mask, VisionFrame
 AgeModifierInputs = TypedDict('AgeModifierInputs',
 {
 	'reference_vision_frame' : VisionFrame,
-	'target_vision_frame' : VisionFrame,
+	'source_vision_frames' : List[VisionFrame],
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })
 
-AgeModifierModel = Literal['styleganex_age']
+AgeModifierModel = Literal['fran', 'styleganex_age']
 
 AgeModifierDirection : TypeAlias = NDArray[Any]

facefusion/processors/modules/background_remover/choices.py

@@ -1,8 +1,8 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_int_range
 from facefusion.processors.modules.background_remover.types import BackgroundRemoverModel
 
-background_remover_models : List[BackgroundRemoverModel] = [ 'ben_2', 'birefnet_general', 'birefnet_portrait', 'isnet_general', 'modnet', 'ormbg', 'rmbg_1.4', 'rmbg_2.0', 'silueta', 'u2net_cloth', 'u2net_general', 'u2net_human', 'u2netp' ]
+background_remover_models : List[BackgroundRemoverModel] = list(get_args(BackgroundRemoverModel))
 
 background_remover_color_range : Sequence[int] = create_int_range(0, 255, 1)

facefusion/processors/modules/background_remover/core.py

@@ -1,14 +1,16 @@
 from argparse import ArgumentParser
 from functools import lru_cache, partial
+from types import ModuleType
 from typing import List, Tuple
 
 import cv2
 import numpy
 
+import facefusion.choices
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, inference_manager, logger, state_manager, translator, video_manager
-from facefusion.common_helper import is_macos
+from facefusion.common_helper import is_macos, is_windows
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
 from facefusion.execution import has_execution_provider
 from facefusion.filesystem import in_directory, is_image, is_video, resolve_relative_path, same_file_extension
@@ -19,8 +21,8 @@ from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.sanitizer import sanitize_int_range
 from facefusion.thread_helper import thread_semaphore
-from facefusion.types import ApplyStateItem, Args, DownloadScope, ExecutionProvider, InferencePool, Mask, ModelOptions, ModelSet, ProcessMode, VisionFrame
+from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, InferenceProvider, Mask, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import read_static_image, read_static_video_frame
+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
@@ -51,6 +53,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/ben_2.onnx')
 				}
 			},
+			'type': 'ben',
 			'size': (1024, 1024),
 			'mean': [ 0.0, 0.0, 0.0 ],
 			'standard_deviation': [ 1.0, 1.0, 1.0 ]
@@ -79,6 +82,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/birefnet_general.onnx')
 				}
 			},
+			'type': 'birefnet',
 			'size': (1024, 1024),
 			'mean': [ 0.0, 0.0, 0.0 ],
 			'standard_deviation': [ 1.0, 1.0, 1.0 ]
@@ -107,10 +111,69 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/birefnet_portrait.onnx')
 				}
 			},
+			'type': 'birefnet',
 			'size': (1024, 1024),
 			'mean': [ 0.0, 0.0, 0.0 ],
 			'standard_deviation': [ 1.0, 1.0, 1.0 ]
 		},
+		'corridor_key_1024':
+		{
+			'__metadata__':
+			{
+				'vendor': 'nikopueringer',
+				'license': 'Non-Commercial',
+				'year': 2025
+			},
+			'hashes':
+			{
+				'background_remover':
+				{
+					'url': resolve_download_url('models-3.6.0', 'corridor_key_1024.hash'),
+					'path': resolve_relative_path('../.assets/models/corridor_key_1024.hash')
+				}
+			},
+			'sources':
+			{
+				'background_remover':
+				{
+					'url': resolve_download_url('models-3.6.0', 'corridor_key_1024.onnx'),
+					'path': resolve_relative_path('../.assets/models/corridor_key_1024.onnx')
+				}
+			},
+			'type': 'corridor_key',
+			'size': (1024, 1024),
+			'mean': [ 0.485, 0.456, 0.406 ],
+			'standard_deviation': [ 0.229, 0.224, 0.225 ]
+		},
+		'corridor_key_2048':
+		{
+			'__metadata__':
+			{
+				'vendor': 'nikopueringer',
+				'license': 'Non-Commercial',
+				'year': 2025
+			},
+			'hashes':
+			{
+				'background_remover':
+				{
+					'url': resolve_download_url('models-3.6.0', 'corridor_key_2048.hash'),
+					'path': resolve_relative_path('../.assets/models/corridor_key_2048.hash')
+				}
+			},
+			'sources':
+			{
+				'background_remover':
+				{
+					'url': resolve_download_url('models-3.6.0', 'corridor_key_2048.onnx'),
+					'path': resolve_relative_path('../.assets/models/corridor_key_2048.onnx')
+				}
+			},
+			'type': 'corridor_key',
+			'size': (2048, 2048),
+			'mean': [ 0.485, 0.456, 0.406 ],
+			'standard_deviation': [ 0.229, 0.224, 0.225 ]
+		},
 		'isnet_general':
 		{
 			'__metadata__':
@@ -135,6 +198,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/isnet_general.onnx')
 				}
 			},
+			'type': 'isnet',
 			'size': (1024, 1024),
 			'mean': [ 0.5, 0.5, 0.5 ],
 			'standard_deviation': [ 1.0, 1.0, 1.0 ]
@@ -163,6 +227,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/modnet.onnx')
 				}
 			},
+			'type': 'modnet',
 			'size': (512, 512),
 			'mean': [ 0.5, 0.5, 0.5 ],
 			'standard_deviation': [ 0.5, 0.5, 0.5 ]
@@ -191,6 +256,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/ormbg.onnx')
 				}
 			},
+			'type': 'ormbg',
 			'size': (1024, 1024),
 			'mean': [ 0.0, 0.0, 0.0 ],
 			'standard_deviation': [ 1.0, 1.0, 1.0 ]
@@ -219,6 +285,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/rmbg_1.4.onnx')
 				}
 			},
+			'type': 'rmbg',
 			'size': (1024, 1024),
 			'mean': [ 0.5, 0.5, 0.5 ],
 			'standard_deviation': [ 1.0, 1.0, 1.0 ]
@@ -247,6 +314,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/rmbg_2.0.onnx')
 				}
 			},
+			'type': 'rmbg',
 			'size': (1024, 1024),
 			'mean': [ 0.485, 0.456, 0.406 ],
 			'standard_deviation': [ 0.229, 0.224, 0.225 ]
@@ -275,6 +343,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/silueta.onnx')
 				}
 			},
+			'type': 'silueta',
 			'size': (320, 320),
 			'mean': [ 0.485, 0.456, 0.406 ],
 			'standard_deviation': [ 0.229, 0.224, 0.225 ]
@@ -303,6 +372,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/u2net_cloth.onnx')
 				}
 			},
+			'type': 'u2net_cloth',
 			'size': (768, 768),
 			'mean': [ 0.485, 0.456, 0.406 ],
 			'standard_deviation': [ 0.229, 0.224, 0.225 ]
@@ -331,6 +401,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/u2net_general.onnx')
 				}
 			},
+			'type': 'u2net',
 			'size': (320, 320),
 			'mean': [ 0.485, 0.456, 0.406 ],
 			'standard_deviation': [ 0.229, 0.224, 0.225 ]
@@ -359,6 +430,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/u2net_human.onnx')
 				}
 			},
+			'type': 'u2net',
 			'size': (320, 320),
 			'mean': [ 0.485, 0.456, 0.406 ],
 			'standard_deviation': [ 0.229, 0.224, 0.225 ]
@@ -387,6 +459,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/u2netp.onnx')
 				}
 			},
+			'type': 'u2netp',
 			'size': (320, 320),
 			'mean': [ 0.485, 0.456, 0.406 ],
 			'standard_deviation': [ 0.229, 0.224, 0.225 ]
@@ -406,10 +479,13 @@ def clear_inference_pool() -> None:
 	inference_manager.clear_inference_pool(__name__, model_names)
 
 
-def resolve_execution_providers() -> List[ExecutionProvider]:
+def resolve_inference_providers() -> List[InferenceProvider]:
-	if is_macos() and has_execution_provider('coreml'):
+	model_type = get_model_options().get('type')
-		return [ 'cpu' ]
+
-	return state_manager.get_item('execution_providers')
+	if is_macos() and has_execution_provider('coreml') or is_windows() and has_execution_provider('directml') and model_type == 'corridor_key':
+		return [ facefusion.choices.execution_provider_set.get('cpu') ]
+
+	return []
 
 
 def get_model_options() -> ModelOptions:
@@ -420,20 +496,30 @@ def get_model_options() -> ModelOptions:
 def register_args(program : ArgumentParser) -> None:
 	group_processors = find_argument_group(program, 'processors')
 	if group_processors:
-		group_processors.add_argument('--background-remover-model', help = translator.get('help.model', __package__), default = config.get_str_value('processors', 'background_remover_model', 'rmbg_2.0'), choices = background_remover_choices.background_remover_models)
+		group_processors.add_argument('--background-remover-model', help = translator.get('help.model', __package__), default = config.get_str_value('processors', 'background_remover_model', 'modnet'), choices = background_remover_choices.background_remover_models)
-		group_processors.add_argument('--background-remover-color', help = translator.get('help.color', __package__), type = partial(sanitize_int_range, int_range = background_remover_choices.background_remover_color_range), default = config.get_int_list('processors', 'background_remover_color', '0 0 0 0'), nargs ='+')
+		group_processors.add_argument('--background-remover-fill-color', help = translator.get('help.fill_color', __package__), type = partial(sanitize_int_range, int_range = background_remover_choices.background_remover_color_range), default = config.get_int_list('processors', 'background_remover_fill_color', '0 0 0 0'), nargs = '+')
-		facefusion.jobs.job_store.register_step_keys([ 'background_remover_model', 'background_remover_color' ])
+		group_processors.add_argument('--background-remover-despill-color', help = translator.get('help.despill_color', __package__), type = partial(sanitize_int_range, int_range = background_remover_choices.background_remover_color_range), default = config.get_int_list('processors', 'background_remover_despill_color', '0 0 0 0'), nargs = '+')
+		facefusion.jobs.job_store.register_step_keys([ 'background_remover_model', 'background_remover_fill_color', 'background_remover_despill_color' ])
 
 
 def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('background_remover_model', args.get('background_remover_model'))
-	apply_state_item('background_remover_color', normalize_color(args.get('background_remover_color')))
+	apply_state_item('background_remover_fill_color', normalize_color(args.get('background_remover_fill_color')))
+	apply_state_item('background_remover_despill_color', normalize_color(args.get('background_remover_despill_color')))
+
+
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser ]
 
 
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -453,24 +539,38 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
+			common_module.clear_inference_pool()
 
 
 def remove_background(temp_vision_frame : VisionFrame) -> Tuple[VisionFrame, Mask]:
-	temp_vision_mask = forward(prepare_temp_frame(temp_vision_frame))
+	model_type = get_model_options().get('type')
-	temp_vision_mask = normalize_vision_mask(temp_vision_mask)
+
-	temp_vision_mask = cv2.resize(temp_vision_mask, temp_vision_frame.shape[:2][::-1])
+	if model_type == 'corridor_key':
-	temp_vision_frame = apply_background_color(temp_vision_frame, temp_vision_mask)
+		remove_vision_mask, remove_vision_frame = forward_corridor_key(prepare_temp_frame(temp_vision_frame))
-	return temp_vision_frame, temp_vision_mask
+		remove_vision_frame = numpy.squeeze(remove_vision_frame).transpose(1, 2, 0)
+		remove_vision_frame = numpy.clip(remove_vision_frame * 255, 0, 255).astype(numpy.uint8)
+		temp_vision_frame = cv2.resize(remove_vision_frame[:, :, ::-1], temp_vision_frame.shape[:2][::-1])
+	else:
+		remove_vision_mask = forward(prepare_temp_frame(temp_vision_frame))
+
+	remove_vision_mask = normalize_vision_mask(remove_vision_mask)
+	remove_vision_mask = cv2.resize(remove_vision_mask, temp_vision_frame.shape[:2][::-1])
+	temp_vision_frame = apply_despill_color(temp_vision_frame)
+	temp_vision_frame = apply_fill_color(temp_vision_frame, remove_vision_mask)
+	return temp_vision_frame, remove_vision_mask
 
 
 def forward(temp_vision_frame : VisionFrame) -> VisionFrame:
 	background_remover = get_inference_pool().get('background_remover')
-	model_name = state_manager.get_item('background_remover_model')
+	model_type = get_model_options().get('type')
 
 	with thread_semaphore():
 		remove_vision_frame = background_remover.run(None,
@@ -478,20 +578,42 @@ def forward(temp_vision_frame : VisionFrame) -> VisionFrame:
 			'input': temp_vision_frame
 		})[0]
 
-		if model_name == 'u2net_cloth':
+		if model_type == 'u2net_cloth':
 			remove_vision_frame = numpy.argmax(remove_vision_frame, axis = 1)
 
 	return remove_vision_frame
 
 
+def forward_corridor_key(temp_vision_frame : VisionFrame) -> Tuple[Mask, VisionFrame]:
+	background_remover = get_inference_pool().get('background_remover')
+
+	with thread_semaphore():
+		remove_vision_mask, remove_vision_frame = background_remover.run(None,
+		{
+			'input': temp_vision_frame
+		})
+
+	return remove_vision_mask, remove_vision_frame
+
+
 def prepare_temp_frame(temp_vision_frame : VisionFrame) -> VisionFrame:
+	model_type = get_model_options().get('type')
 	model_size = get_model_options().get('size')
 	model_mean = get_model_options().get('mean')
 	model_standard_deviation = get_model_options().get('standard_deviation')
 
+	if model_type == 'corridor_key':
+		coarse_color = temp_vision_frame[:, :, ::-1].astype(numpy.float32) / 255.0
+		coarse_bias = coarse_color[:, :, 1] - numpy.maximum(coarse_color[:, :, 0], coarse_color[:, :, 2])
+		coarse_vision_mask = cv2.resize(1.0 - numpy.clip(coarse_bias * 2.0, 0, 1), model_size)[:, :, numpy.newaxis]
+
 	temp_vision_frame = cv2.resize(temp_vision_frame, model_size)
 	temp_vision_frame = temp_vision_frame[:, :, ::-1] / 255.0
 	temp_vision_frame = (temp_vision_frame - model_mean) / model_standard_deviation
+
+	if model_type == 'corridor_key':
+		temp_vision_frame = numpy.concatenate([ temp_vision_frame, coarse_vision_mask ], axis = 2)
+
 	temp_vision_frame = temp_vision_frame.transpose(2, 0, 1)
 	temp_vision_frame = numpy.expand_dims(temp_vision_frame, axis = 0).astype(numpy.float32)
 	return temp_vision_frame
@@ -503,16 +625,32 @@ def normalize_vision_mask(temp_vision_mask : Mask) -> Mask:
 	return temp_vision_mask
 
 
-def apply_background_color(temp_vision_frame : VisionFrame, temp_vision_mask : Mask) -> VisionFrame:
+def apply_fill_color(temp_vision_frame : VisionFrame, temp_vision_mask : Mask) -> VisionFrame:
-	background_remover_color = state_manager.get_item('background_remover_color')
+	background_remover_fill_color = state_manager.get_item('background_remover_fill_color')
 	temp_vision_mask = temp_vision_mask.astype(numpy.float32) / 255
 	temp_vision_mask = numpy.expand_dims(temp_vision_mask, axis = 2)
-	temp_vision_mask = (1 - temp_vision_mask) * background_remover_color[-1] / 255
+	temp_vision_mask = (1 - temp_vision_mask) * background_remover_fill_color[-1] / 255
-	color_frame = numpy.zeros_like(temp_vision_frame)
+	fill_vision_frame = numpy.zeros_like(temp_vision_frame)
-	color_frame[:, :, 0] = background_remover_color[2]
+	fill_vision_frame[:, :, 0] = background_remover_fill_color[2]
-	color_frame[:, :, 1] = background_remover_color[1]
+	fill_vision_frame[:, :, 1] = background_remover_fill_color[1]
-	color_frame[:, :, 2] = background_remover_color[0]
+	fill_vision_frame[:, :, 2] = background_remover_fill_color[0]
-	temp_vision_frame = temp_vision_frame * (1 - temp_vision_mask) + color_frame * temp_vision_mask
+	temp_vision_frame = temp_vision_frame * (1 - temp_vision_mask) + fill_vision_frame * temp_vision_mask
+	temp_vision_frame = temp_vision_frame.astype(numpy.uint8)
+	return temp_vision_frame
+
+
+def apply_despill_color(temp_vision_frame : VisionFrame) -> VisionFrame:
+	background_remover_despill_color = state_manager.get_item('background_remover_despill_color')
+	temp_vision_frame = temp_vision_frame.astype(numpy.float32)
+	color_alpha = background_remover_despill_color[3] / 255.0
+	despill_vision_frame = numpy.zeros_like(temp_vision_frame)
+	despill_vision_frame[:, :, 0] = background_remover_despill_color[2]
+	despill_vision_frame[:, :, 1] = background_remover_despill_color[1]
+	despill_vision_frame[:, :, 2] = background_remover_despill_color[0]
+	color_weight = despill_vision_frame / numpy.maximum(numpy.max(background_remover_despill_color[:3]), 1)
+	color_limit = numpy.roll(temp_vision_frame, 1, 2) + numpy.roll(temp_vision_frame, -1, 2)
+	limit_vision_frame = numpy.minimum(temp_vision_frame, color_limit * 0.5)
+	temp_vision_frame = temp_vision_frame + (limit_vision_frame - temp_vision_frame) * color_alpha * color_weight
 	temp_vision_frame = temp_vision_frame.astype(numpy.uint8)
 	return temp_vision_frame
 

facefusion/processors/modules/background_remover/locales.py

@@ -0,0 +1,26 @@
+from facefusion.types import Locales
+
+LOCALES : Locales =\
+{
+	'en':
+	{
+		'help':
+		{
+			'model': 'choose the model responsible for removing the background',
+			'fill_color': 'apply red, green, blue and alpha values to the background',
+			'despill_color': 'remove red, green, blue and alpha values from the foreground'
+		},
+		'uis':
+		{
+			'model_dropdown': 'BACKGROUND REMOVER MODEL',
+			'fill_color_red_number': 'FILL COLOR RED',
+			'fill_color_green_number': 'FILL COLOR GREEN',
+			'fill_color_blue_number': 'FILL COLOR BLUE',
+			'fill_color_alpha_number': 'FILL COLOR ALPHA',
+			'despill_color_red_number': 'DESPILL COLOR RED',
+			'despill_color_green_number': 'DESPILL COLOR GREEN',
+			'despill_color_blue_number': 'DESPILL COLOR BLUE',
+			'despill_color_alpha_number': 'DESPILL COLOR ALPHA'
+		}
+	}
+}

facefusion/processors/modules/background_remover/locals.py

@@ -1,21 +0,0 @@
-from facefusion.types import Locals
-
-LOCALS : Locals =\
-{
-	'en':
-	{
-		'help':
-		{
-			'model': 'choose the model responsible for removing the background',
-			'color': 'apply red, green blue and alpha values to the background'
-		},
-		'uis':
-		{
-			'model_dropdown': 'BACKGROUND REMOVER MODEL',
-			'color_red_number': 'BACKGROUND COLOR RED',
-			'color_green_number': 'BACKGROUND COLOR GREEN',
-			'color_blue_number': 'BACKGROUND COLOR BLUE',
-			'color_alpha_number': 'BACKGROUND COLOR ALPHA'
-		}
-	}
-}

facefusion/processors/modules/background_remover/types.py

@@ -1,12 +1,12 @@
-from typing import Literal, TypedDict
+from typing import List, Literal, TypedDict
 
 from facefusion.types import Mask, VisionFrame
 
 BackgroundRemoverInputs = TypedDict('BackgroundRemoverInputs',
 {
-	'target_vision_frame' : VisionFrame,
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })
 
-BackgroundRemoverModel = Literal['ben_2', 'birefnet_general', 'birefnet_portrait', 'isnet_general', 'modnet', 'ormbg', 'rmbg_1.4', 'rmbg_2.0', 'silueta', 'u2net_cloth', 'u2net_general', 'u2net_human', 'u2netp']
+BackgroundRemoverModel = Literal['ben_2', 'birefnet_general', 'birefnet_portrait', 'corridor_key_1024', 'corridor_key_2048', 'isnet_general', 'modnet', 'ormbg', 'rmbg_1.4', 'rmbg_2.0', 'silueta', 'u2net_cloth', 'u2net_general', 'u2net_human', 'u2netp']

facefusion/processors/modules/deep_swapper/core.py

@@ -1,6 +1,7 @@
 from argparse import ArgumentParser
 from functools import lru_cache
-from typing import Tuple
+from types import ModuleType
+from typing import List, Tuple
 
 import cv2
 import numpy
@@ -9,9 +10,9 @@ from cv2.typing import Size
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager
-from facefusion.common_helper import create_int_metavar
+from facefusion.common_helper import create_int_metavar, get_middle
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url_by_provider
-from facefusion.face_analyser import scale_face
+from facefusion.face_creator import scale_face
 from facefusion.face_helper import paste_back, warp_face_by_face_landmark_5
 from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_mask, create_region_mask
 from facefusion.face_selector import select_faces
@@ -22,7 +23,7 @@ from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import thread_semaphore
 from facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, Mask, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import conditional_match_frame_color, read_static_image, read_static_video_frame
+from facefusion.vision import conditional_match_frame_color, read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
@@ -286,10 +287,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('deep_swapper_morph', args.get('deep_swapper_morph'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	if model_hash_set and model_source_set:
 		return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 	return True
@@ -311,16 +320,15 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
 
 
 def swap_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
@@ -411,10 +419,13 @@ def prepare_crop_mask(crop_source_mask : Mask, crop_target_mask : Mask) -> Mask:
 
 def process_frame(inputs : DeepSwapperInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
-	target_vision_frame = inputs.get('target_vision_frame')
+	source_vision_frames = inputs.get('source_vision_frames')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+
+	target_vision_frame = get_middle(target_vision_frames)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if target_faces:
 		for target_face in target_faces:

facefusion/processors/modules/deep_swapper/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/deep_swapper/types.py

@@ -1,4 +1,4 @@
-from typing import Any, TypeAlias, TypedDict
+from typing import Any, List, TypeAlias, TypedDict
 
 from numpy.typing import NDArray
 
@@ -7,7 +7,8 @@ from facefusion.types import Mask, VisionFrame
 DeepSwapperInputs = TypedDict('DeepSwapperInputs',
 {
 	'reference_vision_frame' : VisionFrame,
-	'target_vision_frame' : VisionFrame,
+	'source_vision_frames' : List[VisionFrame],
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/expression_restorer/choices.py

@@ -1,10 +1,10 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_int_range
 from facefusion.processors.modules.expression_restorer.types import ExpressionRestorerArea, ExpressionRestorerModel
 
-expression_restorer_models : List[ExpressionRestorerModel] = [ 'live_portrait' ]
+expression_restorer_models : List[ExpressionRestorerModel] = list(get_args(ExpressionRestorerModel))
 
-expression_restorer_areas : List[ExpressionRestorerArea] = [ 'upper-face', 'lower-face' ]
+expression_restorer_areas : List[ExpressionRestorerArea] = list(get_args(ExpressionRestorerArea))
 
 expression_restorer_factor_range : Sequence[int] = create_int_range(0, 100, 1)

facefusion/processors/modules/expression_restorer/core.py

@@ -1,6 +1,7 @@
 from argparse import ArgumentParser
 from functools import lru_cache
-from typing import Tuple
+from types import ModuleType
+from typing import List, Tuple
 
 import cv2
 import numpy
@@ -8,9 +9,9 @@ import numpy
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager
-from facefusion.common_helper import create_int_metavar
+from facefusion.common_helper import create_int_metavar, get_middle
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
-from facefusion.face_analyser import scale_face
+from facefusion.face_creator import scale_face
 from facefusion.face_helper import paste_back, warp_face_by_face_landmark_5
 from facefusion.face_masker import create_box_mask, create_occlusion_mask
 from facefusion.face_selector import select_faces
@@ -22,7 +23,7 @@ from facefusion.processors.types import LivePortraitExpression, LivePortraitFeat
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import conditional_thread_semaphore, thread_semaphore
 from facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import read_static_image, read_static_video_frame
+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
@@ -101,7 +102,7 @@ def register_args(program : ArgumentParser) -> None:
 	if group_processors:
 		group_processors.add_argument('--expression-restorer-model', help = translator.get('help.model', __package__), default = config.get_str_value('processors', 'expression_restorer_model', 'live_portrait'), choices = expression_restorer_choices.expression_restorer_models)
 		group_processors.add_argument('--expression-restorer-factor', help = translator.get('help.factor', __package__), type = int, default = config.get_int_value('processors', 'expression_restorer_factor', '80'), choices = expression_restorer_choices.expression_restorer_factor_range, metavar = create_int_metavar(expression_restorer_choices.expression_restorer_factor_range))
-		group_processors.add_argument('--expression-restorer-areas', help = translator.get('help.areas', __package__).format(choices = ', '.join(expression_restorer_choices.expression_restorer_areas)), default = config.get_str_list('processors', 'expression_restorer_areas', ' '.join(expression_restorer_choices.expression_restorer_areas)), choices = expression_restorer_choices.expression_restorer_areas, nargs ='+', metavar ='EXPRESSION_RESTORER_AREAS')
+		group_processors.add_argument('--expression-restorer-areas', help = translator.get('help.areas', __package__).format(choices = ', '.join(expression_restorer_choices.expression_restorer_areas)), default = config.get_str_list('processors', 'expression_restorer_areas', ' '.join(expression_restorer_choices.expression_restorer_areas)), choices = expression_restorer_choices.expression_restorer_areas, nargs = '+', metavar = 'EXPRESSION_RESTORER_AREAS')
 		facefusion.jobs.job_store.register_step_keys([ 'expression_restorer_model', 'expression_restorer_factor', 'expression_restorer_areas' ])
 
 
@@ -111,10 +112,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('expression_restorer_areas', args.get('expression_restorer_areas'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -137,16 +146,15 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
 
 
 def restore_expression(target_face : Face, target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> VisionFrame:
@@ -192,12 +200,12 @@ def restrict_expression_areas(temp_expression : LivePortraitExpression, target_e
 	expression_restorer_areas = state_manager.get_item('expression_restorer_areas')
 
 	if 'upper-face' not in expression_restorer_areas:
-		target_expression[:, [1, 2, 6, 10, 11, 12, 13, 15, 16]] = temp_expression[:, [1, 2, 6, 10, 11, 12, 13, 15, 16]]
+		target_expression[:, [ 1, 2, 6, 10, 11, 12, 13, 15, 16 ]] = temp_expression[:, [ 1, 2, 6, 10, 11, 12, 13, 15, 16 ]]
 
 	if 'lower-face' not in expression_restorer_areas:
-		target_expression[:, [3, 7, 14, 17, 18, 19, 20]] = temp_expression[:, [3, 7, 14, 17, 18, 19, 20]]
+		target_expression[:, [ 3, 7, 14, 17, 18, 19, 20 ]] = temp_expression[:, [ 3, 7, 14, 17, 18, 19, 20 ]]
 
-	target_expression[:, [0, 4, 5, 8, 9]] = temp_expression[:, [0, 4, 5, 8, 9]]
+	target_expression[:, [ 0, 4, 5, 8, 9 ]] = temp_expression[:, [ 0, 4, 5, 8, 9 ]]
 	return target_expression
 
 
@@ -257,10 +265,13 @@ def normalize_crop_frame(crop_vision_frame : VisionFrame) -> VisionFrame:
 
 def process_frame(inputs : ExpressionRestorerInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
-	target_vision_frame = inputs.get('target_vision_frame')
+	source_vision_frames = inputs.get('source_vision_frames')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+
+	target_vision_frame = get_middle(target_vision_frames)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if target_faces:
 		for target_face in target_faces:

facefusion/processors/modules/expression_restorer/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/expression_restorer/types.py

@@ -6,7 +6,7 @@ ExpressionRestorerInputs = TypedDict('ExpressionRestorerInputs',
 {
 	'reference_vision_frame' : VisionFrame,
 	'source_vision_frames' : List[VisionFrame],
-	'target_vision_frame' : VisionFrame,
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/face_debugger/choices.py

@@ -1,5 +1,5 @@
-from typing import List
+from typing import List, get_args
 
 from facefusion.processors.modules.face_debugger.types import FaceDebuggerItem
 
-face_debugger_items : List[FaceDebuggerItem] = [ 'bounding-box', 'face-landmark-5', 'face-landmark-5/68', 'face-landmark-68', 'face-landmark-68/5', 'face-mask' ]
+face_debugger_items : List[FaceDebuggerItem] = list(get_args(FaceDebuggerItem))

facefusion/processors/modules/face_debugger/core.py

@@ -1,4 +1,6 @@
 from argparse import ArgumentParser
+from types import ModuleType
+from typing import List
 
 import cv2
 import numpy
@@ -6,7 +8,8 @@ import numpy
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, logger, state_manager, translator, video_manager
-from facefusion.face_analyser import scale_face
+from facefusion.common_helper import get_middle
+from facefusion.face_creator import scale_face
 from facefusion.face_helper import warp_face_by_face_landmark_5
 from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_mask, create_region_mask
 from facefusion.face_selector import select_faces
@@ -16,7 +19,7 @@ from facefusion.processors.modules.face_debugger.types import FaceDebuggerInputs
 from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.types import ApplyStateItem, Args, Face, InferencePool, ProcessMode, VisionFrame
-from facefusion.vision import read_static_image, read_static_video_frame
+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 def get_inference_pool() -> InferencePool:
@@ -38,7 +41,14 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('face_debugger_items', args.get('face_debugger_items'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]
+
+
 def pre_check() -> bool:
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
 	return True
 
 
@@ -58,14 +68,12 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
 
 
 def debug_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
@@ -94,21 +102,22 @@ def debug_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFra
 
 def draw_bounding_box(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
 	temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)
-	box_color = 0, 0, 255
-	border_color = 100, 100, 255
 	bounding_box = target_face.bounding_box.astype(numpy.int32)
 	x1, y1, x2, y2 = bounding_box
+	box_color = 0, 0, 255
+	border_scale = calculate_scale(temp_vision_frame)
+	border_color = 100, 100, 255
 
-	cv2.rectangle(temp_vision_frame, (x1, y1), (x2, y2), box_color, 2)
+	cv2.rectangle(temp_vision_frame, (x1, y1), (x2, y2), box_color, border_scale)
 
 	if target_face.angle == 0:
-		cv2.line(temp_vision_frame, (x1, y1), (x2, y1), border_color, 3)
+		cv2.line(temp_vision_frame, (x1, y1), (x2, y1), border_color, border_scale + 1)
 	if target_face.angle == 180:
-		cv2.line(temp_vision_frame, (x1, y2), (x2, y2), border_color, 3)
+		cv2.line(temp_vision_frame, (x1, y2), (x2, y2), border_color, border_scale + 1)
 	if target_face.angle == 90:
-		cv2.line(temp_vision_frame, (x2, y1), (x2, y2), border_color, 3)
+		cv2.line(temp_vision_frame, (x2, y1), (x2, y2), border_color, border_scale + 1)
 	if target_face.angle == 270:
-		cv2.line(temp_vision_frame, (x1, y1), (x1, y2), border_color, 3)
+		cv2.line(temp_vision_frame, (x1, y1), (x1, y2), border_color, border_scale + 1)
 
 	return temp_vision_frame
 
@@ -122,11 +131,15 @@ def draw_face_mask(target_face : Face, temp_vision_frame : VisionFrame) -> Visio
 	crop_vision_frame, affine_matrix = warp_face_by_face_landmark_5(temp_vision_frame, face_landmark_5_68, 'arcface_128', (512, 512))
 	inverse_matrix = cv2.invertAffineTransform(affine_matrix)
 	temp_size = temp_vision_frame.shape[:2][::-1]
+	mask_scale = calculate_scale(temp_vision_frame)
 	mask_color = 0, 255, 0
 
 	if numpy.array_equal(face_landmark_5, face_landmark_5_68):
 		mask_color = 255, 255, 0
 
+	if target_face.origin == 'refill':
+		mask_color = 0, 165, 255
+
 	if 'box' in state_manager.get_item('face_mask_types'):
 		box_mask = create_box_mask(crop_vision_frame, 0, state_manager.get_item('face_mask_padding'))
 		crop_masks.append(box_mask)
@@ -149,7 +162,7 @@ def draw_face_mask(target_face : Face, temp_vision_frame : VisionFrame) -> Visio
 	inverse_vision_frame = cv2.warpAffine(crop_mask, inverse_matrix, temp_size)
 	inverse_vision_frame = cv2.threshold(inverse_vision_frame, 100, 255, cv2.THRESH_BINARY)[1]
 	inverse_contours, _ = cv2.findContours(inverse_vision_frame, cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE)
-	cv2.drawContours(temp_vision_frame, inverse_contours, -1, mask_color, 2)
+	cv2.drawContours(temp_vision_frame, inverse_contours, -1, mask_color, mask_scale)
 
 	return temp_vision_frame
 
@@ -157,13 +170,17 @@ def draw_face_mask(target_face : Face, temp_vision_frame : VisionFrame) -> Visio
 def draw_face_landmark_5(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
 	temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)
 	face_landmark_5 = target_face.landmark_set.get('5')
+	point_scale = calculate_scale(temp_vision_frame)
 	point_color = 0, 0, 255
 
+	if target_face.origin == 'refill':
+		point_color = 0, 165, 255
+
 	if numpy.any(face_landmark_5):
 		face_landmark_5 = face_landmark_5.astype(numpy.int32)
 
 		for point in face_landmark_5:
-			cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)
+			cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)
 
 	return temp_vision_frame
 
@@ -172,16 +189,20 @@ def draw_face_landmark_5_68(target_face : Face, temp_vision_frame : VisionFrame)
 	temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)
 	face_landmark_5 = target_face.landmark_set.get('5')
 	face_landmark_5_68 = target_face.landmark_set.get('5/68')
+	point_scale = calculate_scale(temp_vision_frame)
 	point_color = 0, 255, 0
 
 	if numpy.array_equal(face_landmark_5, face_landmark_5_68):
 		point_color = 255, 255, 0
 
+	if target_face.origin == 'refill':
+		point_color = 0, 165, 255
+
 	if numpy.any(face_landmark_5_68):
 		face_landmark_5_68 = face_landmark_5_68.astype(numpy.int32)
 
 		for point in face_landmark_5_68:
-			cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)
+			cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)
 
 	return temp_vision_frame
 
@@ -190,16 +211,20 @@ def draw_face_landmark_68(target_face : Face, temp_vision_frame : VisionFrame) -
 	temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)
 	face_landmark_68 = target_face.landmark_set.get('68')
 	face_landmark_68_5 = target_face.landmark_set.get('68/5')
+	point_scale = calculate_scale(temp_vision_frame)
 	point_color = 0, 255, 0
 
 	if numpy.array_equal(face_landmark_68, face_landmark_68_5):
 		point_color = 255, 255, 0
 
+	if target_face.origin == 'refill':
+		point_color = 0, 165, 255
+
 	if numpy.any(face_landmark_68):
 		face_landmark_68 = face_landmark_68.astype(numpy.int32)
 
 		for point in face_landmark_68:
-			cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)
+			cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)
 
 	return temp_vision_frame
 
@@ -207,23 +232,36 @@ def draw_face_landmark_68(target_face : Face, temp_vision_frame : VisionFrame) -
 def draw_face_landmark_68_5(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
 	temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)
 	face_landmark_68_5 = target_face.landmark_set.get('68/5')
+	point_scale = calculate_scale(temp_vision_frame)
 	point_color = 255, 255, 0
 
+	if target_face.origin == 'refill':
+		point_color = 0, 165, 255
+
 	if numpy.any(face_landmark_68_5):
 		face_landmark_68_5 = face_landmark_68_5.astype(numpy.int32)
 
 		for point in face_landmark_68_5:
-			cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)
+			cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)
 
 	return temp_vision_frame
 
 
+def calculate_scale(temp_vision_frame : VisionFrame) -> int:
+	frame_height, _ = temp_vision_frame.shape[:2]
+	frame_scale = round(frame_height / 270)
+	return max(1, min(10, frame_scale))
+
+
 def process_frame(inputs : FaceDebuggerInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
-	target_vision_frame = inputs.get('target_vision_frame')
+	source_vision_frames = inputs.get('source_vision_frames')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+
+	target_vision_frame = get_middle(target_vision_frames)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if target_faces:
 		for target_face in target_faces:
@@ -231,5 +269,3 @@ def process_frame(inputs : FaceDebuggerInputs) -> ProcessorOutputs:
 			temp_vision_frame = debug_face(target_face, temp_vision_frame)
 
 	return temp_vision_frame, temp_vision_mask
-
-

facefusion/processors/modules/face_debugger/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/face_debugger/types.py

@@ -1,11 +1,12 @@
-from typing import Literal, TypedDict
+from typing import List, Literal, TypedDict
 
 from facefusion.types import Mask, VisionFrame
 
 FaceDebuggerInputs = TypedDict('FaceDebuggerInputs',
 {
 	'reference_vision_frame' : VisionFrame,
-	'target_vision_frame' : VisionFrame,
+	'source_vision_frames' : List[VisionFrame],
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/face_editor/choices.py

@@ -1,9 +1,9 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_float_range
 from facefusion.processors.modules.face_editor.types import FaceEditorModel
 
-face_editor_models : List[FaceEditorModel] = [ 'live_portrait' ]
+face_editor_models : List[FaceEditorModel] = list(get_args(FaceEditorModel))
 
 face_editor_eyebrow_direction_range : Sequence[float] = create_float_range(-1.0, 1.0, 0.05)
 face_editor_eye_gaze_horizontal_range : Sequence[float] = create_float_range(-1.0, 1.0, 0.05)

facefusion/processors/modules/face_editor/core.py

@@ -1,6 +1,7 @@
 from argparse import ArgumentParser
 from functools import lru_cache
-from typing import Tuple
+from types import ModuleType
+from typing import List, Tuple
 
 import cv2
 import numpy
@@ -8,9 +9,9 @@ import numpy
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager
-from facefusion.common_helper import create_float_metavar
+from facefusion.common_helper import create_float_metavar, get_middle
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
-from facefusion.face_analyser import scale_face
+from facefusion.face_creator import scale_face
 from facefusion.face_helper import paste_back, scale_face_landmark_5, warp_face_by_face_landmark_5
 from facefusion.face_masker import create_box_mask
 from facefusion.face_selector import select_faces
@@ -22,7 +23,7 @@ from facefusion.processors.types import LivePortraitExpression, LivePortraitFeat
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import conditional_thread_semaphore, thread_semaphore
 from facefusion.types import ApplyStateItem, Args, DownloadScope, Face, FaceLandmark68, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import read_static_image, read_static_video_frame
+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
@@ -165,10 +166,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('face_editor_head_roll', args.get('face_editor_head_roll'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -188,16 +197,15 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
 
 
 def edit_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
@@ -486,10 +494,13 @@ def normalize_crop_frame(crop_vision_frame : VisionFrame) -> VisionFrame:
 
 def process_frame(inputs : FaceEditorInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
-	target_vision_frame = inputs.get('target_vision_frame')
+	source_vision_frames = inputs.get('source_vision_frames')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+
+	target_vision_frame = get_middle(target_vision_frames)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if target_faces:
 		for target_face in target_faces:

facefusion/processors/modules/face_editor/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/face_editor/types.py

@@ -1,11 +1,12 @@
-from typing import Literal, TypedDict
+from typing import List, Literal, TypedDict
 
 from facefusion.types import Mask, VisionFrame
 
 FaceEditorInputs = TypedDict('FaceEditorInputs',
 {
 	'reference_vision_frame' : VisionFrame,
-	'target_vision_frame' : VisionFrame,
+	'source_vision_frames' : List[VisionFrame],
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/face_enhancer/choices.py

@@ -1,9 +1,9 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_float_range, create_int_range
 from facefusion.processors.modules.face_enhancer.types import FaceEnhancerModel
 
-face_enhancer_models : List[FaceEnhancerModel] = [ 'codeformer', 'gfpgan_1.2', 'gfpgan_1.3', 'gfpgan_1.4', 'gpen_bfr_256', 'gpen_bfr_512', 'gpen_bfr_1024', 'gpen_bfr_2048', 'restoreformer_plus_plus' ]
+face_enhancer_models : List[FaceEnhancerModel] = list(get_args(FaceEnhancerModel))
 
 face_enhancer_blend_range : Sequence[int] = create_int_range(0, 100, 1)
 

facefusion/processors/modules/face_enhancer/core.py

@@ -1,14 +1,16 @@
 from argparse import ArgumentParser
 from functools import lru_cache
+from types import ModuleType
+from typing import List
 
 import numpy
 
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager
-from facefusion.common_helper import create_float_metavar, create_int_metavar
+from facefusion.common_helper import create_float_metavar, create_int_metavar, get_middle
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
-from facefusion.face_analyser import scale_face
+from facefusion.face_creator import scale_face
 from facefusion.face_helper import paste_back, warp_face_by_face_landmark_5
 from facefusion.face_masker import create_box_mask, create_occlusion_mask
 from facefusion.face_selector import select_faces
@@ -19,7 +21,7 @@ from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import thread_semaphore
 from facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import blend_frame, read_static_image, read_static_video_frame
+from facefusion.vision import blend_frame, read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
@@ -304,10 +306,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('face_enhancer_weight', args.get('face_enhancer_weight'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -327,16 +337,15 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
 
 
 def enhance_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
@@ -413,10 +422,13 @@ def blend_paste_frame(temp_vision_frame : VisionFrame, paste_vision_frame : Visi
 
 def process_frame(inputs : FaceEnhancerInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
-	target_vision_frame = inputs.get('target_vision_frame')
+	source_vision_frames = inputs.get('source_vision_frames')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+
+	target_vision_frame = get_middle(target_vision_frames)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if target_faces:
 		for target_face in target_faces:

facefusion/processors/modules/face_enhancer/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/face_enhancer/types.py

@@ -1,4 +1,4 @@
-from typing import Any, Literal, TypeAlias, TypedDict
+from typing import Any, List, Literal, TypeAlias, TypedDict
 
 from numpy.typing import NDArray
 
@@ -7,7 +7,8 @@ from facefusion.types import Mask, VisionFrame
 FaceEnhancerInputs = TypedDict('FaceEnhancerInputs',
 {
 	'reference_vision_frame' : VisionFrame,
-	'target_vision_frame' : VisionFrame,
+	'source_vision_frames' : List[VisionFrame],
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/face_swapper/choices.py

@@ -1,8 +1,9 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_float_range
 from facefusion.processors.modules.face_swapper.types import FaceSwapperModel, FaceSwapperSet, FaceSwapperWeight
 
+
 face_swapper_set : FaceSwapperSet =\
 {
 	'blendswap_256': [ '256x256', '384x384', '512x512', '768x768', '1024x1024' ],
@@ -20,6 +21,6 @@ face_swapper_set : FaceSwapperSet =\
 	'uniface_256': [ '256x256', '512x512', '768x768', '1024x1024' ]
 }
 
-face_swapper_models : List[FaceSwapperModel] = list(face_swapper_set.keys())
+face_swapper_models : List[FaceSwapperModel] = list(get_args(FaceSwapperModel))
 
 face_swapper_weight_range : Sequence[FaceSwapperWeight] = create_float_range(0.0, 1.0, 0.05)

facefusion/processors/modules/face_swapper/core.py

@@ -1,5 +1,6 @@
 from argparse import ArgumentParser
 from functools import lru_cache
+from types import ModuleType
 from typing import List, Optional, Tuple
 
 import cv2
@@ -9,10 +10,10 @@ import facefusion.choices
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager
-from facefusion.common_helper import get_first, is_macos
+from facefusion.common_helper import get_first, get_middle, is_macos
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
 from facefusion.execution import has_execution_provider
-from facefusion.face_analyser import get_average_face, get_many_faces, get_one_face, scale_face
+from facefusion.face_creator import average_face_identity, get_one_face, get_static_faces, scale_face
 from facefusion.face_helper import paste_back, warp_face_by_face_landmark_5
 from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_mask, create_region_mask
 from facefusion.face_selector import select_faces, sort_faces_by_order
@@ -24,8 +25,8 @@ from facefusion.processors.pixel_boost import explode_pixel_boost, implode_pixel
 from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import conditional_thread_semaphore
-from facefusion.types import ApplyStateItem, Args, DownloadScope, Embedding, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
+from facefusion.types import ApplyStateItem, Args, DownloadScope, Embedding, Face, InferencePool, InferenceProvider, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import read_static_image, read_static_images, read_static_video_frame, unpack_resolution
+from facefusion.vision import read_static_image, read_static_images, read_static_video_chunk, read_static_video_frame, unpack_resolution
 
 
 @lru_cache()
@@ -246,6 +247,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/hyperswap_1a_256.onnx')
 				}
 			},
+			'precision': 'fp16',
 			'type': 'hyperswap',
 			'template': 'arcface_128',
 			'size': (256, 256),
@@ -276,6 +278,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 						'path': resolve_relative_path('../.assets/models/hyperswap_1b_256.onnx')
 					}
 				},
+			'precision': 'fp16',
 			'type': 'hyperswap',
 			'template': 'arcface_128',
 			'size': (256, 256),
@@ -306,6 +309,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/hyperswap_1c_256.onnx')
 				}
 			},
+			'precision': 'fp16',
 			'type': 'hyperswap',
 			'template': 'arcface_128',
 			'size': (256, 256),
@@ -366,6 +370,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/inswapper_128_fp16.onnx')
 				}
 			},
+			'precision': 'fp16',
 			'type': 'inswapper',
 			'template': 'arcface_128',
 			'size': (128, 128),
@@ -486,28 +491,38 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 
 
 def get_inference_pool() -> InferencePool:
-	model_names = [ get_model_name() ]
+	model_names = [ state_manager.get_item('face_swapper_model') ]
 	model_source_set = get_model_options().get('sources')
 
 	return inference_manager.get_inference_pool(__name__, model_names, model_source_set)
 
 
 def clear_inference_pool() -> None:
-	model_names = [ get_model_name() ]
+	model_names = [ state_manager.get_item('face_swapper_model') ]
 	inference_manager.clear_inference_pool(__name__, model_names)
 
 
+def resolve_inference_providers() -> List[InferenceProvider]:
+	model_precision = get_model_options().get('precision')
+	model_type = get_model_options().get('type')
+
+	if is_macos() and has_execution_provider('coreml'):
+		if model_type in [ 'ghost', 'uniface' ] or model_precision == 'fp16':
+			return\
+			[
+				(facefusion.choices.execution_provider_set.get('coreml'),
+				{
+					'ModelFormat': 'MLProgram',
+					'SpecializationStrategy': 'FastPrediction'
+				})
+			]
+
+	return []
+
+
 def get_model_options() -> ModelOptions:
-	model_name = get_model_name()
-	return create_static_model_set('full').get(model_name)
-
-
-def get_model_name() -> str:
 	model_name = state_manager.get_item('face_swapper_model')
-
+	return create_static_model_set('full').get(model_name)
-	if is_macos() and has_execution_provider('coreml') and model_name == 'inswapper_128_fp16':
-		return 'inswapper_128'
-	return model_name
 
 
 def register_args(program : ArgumentParser) -> None:
@@ -527,10 +542,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('face_swapper_weight', args.get('face_swapper_weight'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -540,8 +563,8 @@ def pre_process(mode : ProcessMode) -> bool:
 		return False
 
 	source_image_paths = filter_image_paths(state_manager.get_item('source_paths'))
-	source_frames = read_static_images(source_image_paths)
+	source_vision_frames = read_static_images(source_image_paths)
-	source_faces = get_many_faces(source_frames)
+	source_faces = get_static_faces(source_vision_frames)
 
 	if not get_one_face(source_faces):
 		logger.error(translator.get('no_source_face_detected') + translator.get('exclamation_mark'), __name__)
@@ -565,20 +588,19 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		get_static_model_initializer.cache_clear()
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
 
 
-def swap_face(source_face : Face, target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:
+def swap_face(source_face : Face, target_face : Face, source_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> VisionFrame:
 	model_template = get_model_options().get('template')
 	model_size = get_model_options().get('size')
 	pixel_boost_size = unpack_resolution(state_manager.get_item('face_swapper_pixel_boost'))
@@ -598,7 +620,7 @@ def swap_face(source_face : Face, target_face : Face, temp_vision_frame : Vision
 	pixel_boost_vision_frames = implode_pixel_boost(crop_vision_frame, pixel_boost_total, model_size)
 	for pixel_boost_vision_frame in pixel_boost_vision_frames:
 		pixel_boost_vision_frame = prepare_crop_frame(pixel_boost_vision_frame)
-		pixel_boost_vision_frame = forward_swap_face(source_face, target_face, pixel_boost_vision_frame)
+		pixel_boost_vision_frame = forward_swap_face(source_face, target_face, source_vision_frame, pixel_boost_vision_frame)
 		pixel_boost_vision_frame = normalize_crop_frame(pixel_boost_vision_frame)
 		temp_vision_frames.append(pixel_boost_vision_frame)
 	crop_vision_frame = explode_pixel_boost(temp_vision_frames, pixel_boost_total, model_size, pixel_boost_size)
@@ -617,18 +639,15 @@ def swap_face(source_face : Face, target_face : Face, temp_vision_frame : Vision
 	return paste_vision_frame
 
 
-def forward_swap_face(source_face : Face, target_face : Face, crop_vision_frame : VisionFrame) -> VisionFrame:
+def forward_swap_face(source_face : Face, target_face : Face, source_vision_frame : VisionFrame, crop_vision_frame : VisionFrame) -> VisionFrame:
 	face_swapper = get_inference_pool().get('face_swapper')
 	model_type = get_model_options().get('type')
 	face_swapper_inputs = {}
 
-	if is_macos() and has_execution_provider('coreml') and model_type in [ 'ghost', 'uniface' ]:
-		face_swapper.set_providers([ facefusion.choices.execution_provider_set.get('cpu') ])
-
 	for face_swapper_input in face_swapper.get_inputs():
 		if face_swapper_input.name == 'source':
 			if model_type in [ 'blendswap', 'uniface' ]:
-				face_swapper_inputs[face_swapper_input.name] = prepare_source_frame(source_face)
+				face_swapper_inputs[face_swapper_input.name] = prepare_source_frame(source_face, source_vision_frame)
 			else:
 				source_embedding = prepare_source_embedding(source_face)
 				source_embedding = balance_source_embedding(source_embedding, target_face.embedding)
@@ -654,9 +673,8 @@ def forward_convert_embedding(face_embedding : Embedding) -> Embedding:
 	return face_embedding
 
 
-def prepare_source_frame(source_face : Face) -> VisionFrame:
+def prepare_source_frame(source_face : Face, source_vision_frame : VisionFrame) -> VisionFrame:
 	model_type = get_model_options().get('type')
-	source_vision_frame = read_static_image(get_first(state_manager.get_item('source_paths')))
 
 	if model_type == 'blendswap':
 		source_vision_frame, _ = warp_face_by_face_landmark_5(source_vision_frame, source_face.landmark_set.get('5/68'), 'arcface_112_v2', (112, 112))
@@ -748,27 +766,31 @@ def extract_source_face(source_vision_frames : List[VisionFrame]) -> Optional[Fa
 
 	if source_vision_frames:
 		for source_vision_frame in source_vision_frames:
-			temp_faces = get_many_faces([source_vision_frame])
+			temp_faces = get_static_faces([ source_vision_frame ])
 			temp_faces = sort_faces_by_order(temp_faces, 'large-small')
 
 			if temp_faces:
 				source_faces.append(get_first(temp_faces))
 
-	return get_average_face(source_faces)
+	return average_face_identity(source_faces)
 
 
 def process_frame(inputs : FaceSwapperInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
 	source_vision_frames = inputs.get('source_vision_frames')
-	target_vision_frame = inputs.get('target_vision_frame')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
+
+	target_vision_frame = get_middle(target_vision_frames)
 	source_face = extract_source_face(source_vision_frames)
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if source_face and target_faces:
+		source_vision_frame = get_first(source_vision_frames)
+
 		for target_face in target_faces:
 			target_face = scale_face(target_face, target_vision_frame, temp_vision_frame)
-			temp_vision_frame = swap_face(source_face, target_face, temp_vision_frame)
+			temp_vision_frame = swap_face(source_face, target_face, source_vision_frame, temp_vision_frame)
 
 	return temp_vision_frame, temp_vision_mask

facefusion/processors/modules/face_swapper/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/face_swapper/types.py

@@ -6,7 +6,7 @@ FaceSwapperInputs = TypedDict('FaceSwapperInputs',
 {
 	'reference_vision_frame' : VisionFrame,
 	'source_vision_frames' : List[VisionFrame],
-	'target_vision_frame' : VisionFrame,
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/frame_colorizer/choices.py

@@ -1,9 +1,9 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_int_range
 from facefusion.processors.modules.frame_colorizer.types import FrameColorizerModel
 
-frame_colorizer_models : List[FrameColorizerModel] = [ 'ddcolor', 'ddcolor_artistic', 'deoldify', 'deoldify_artistic', 'deoldify_stable' ]
+frame_colorizer_models : List[FrameColorizerModel] = list(get_args(FrameColorizerModel))
 
 frame_colorizer_sizes : List[str] = [ '192x192', '256x256', '384x384', '512x512' ]
 

facefusion/processors/modules/frame_colorizer/core.py

@@ -1,10 +1,12 @@
 from argparse import ArgumentParser
 from functools import lru_cache
+from types import ModuleType
 from typing import List
 
 import cv2
 import numpy
 
+import facefusion.choices
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, inference_manager, logger, state_manager, translator, video_manager
@@ -17,8 +19,8 @@ from facefusion.processors.modules.frame_colorizer.types import FrameColorizerIn
 from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import thread_semaphore
-from facefusion.types import ApplyStateItem, Args, DownloadScope, ExecutionProvider, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
+from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, InferenceProvider, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import blend_frame, read_static_image, read_static_video_frame, unpack_resolution
+from facefusion.vision import blend_frame, read_static_image, read_static_video_chunk, read_static_video_frame, unpack_resolution
 
 
 @lru_cache()
@@ -170,10 +172,11 @@ def clear_inference_pool() -> None:
 	inference_manager.clear_inference_pool(__name__, model_names)
 
 
-def resolve_execution_providers() -> List[ExecutionProvider]:
+def resolve_inference_providers() -> List[InferenceProvider]:
 	if is_macos() and has_execution_provider('coreml'):
-		return [ 'cpu' ]
+		return [ facefusion.choices.execution_provider_set.get('cpu') ]
-	return state_manager.get_item('execution_providers')
+
+	return []
 
 
 def get_model_options() -> ModelOptions:
@@ -196,10 +199,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('frame_colorizer_size', args.get('frame_colorizer_size'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -219,11 +230,15 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
+			common_module.clear_inference_pool()
 
 
 def colorize_frame(temp_vision_frame : VisionFrame) -> VisionFrame:

facefusion/processors/modules/frame_colorizer/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/frame_colorizer/types.py

@@ -1,10 +1,10 @@
-from typing import Literal, TypedDict
+from typing import List, Literal, TypedDict
 
 from facefusion.types import Mask, VisionFrame
 
 FrameColorizerInputs = TypedDict('FrameColorizerInputs',
 {
-	'target_vision_frame' : VisionFrame,
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/frame_enhancer/choices.py

@@ -1,8 +1,8 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_int_range
 from facefusion.processors.modules.frame_enhancer.types import FrameEnhancerModel
 
-frame_enhancer_models : List[FrameEnhancerModel] = [ 'clear_reality_x4', 'face_dat_x4', 'lsdir_x4', 'nomos8k_sc_x4', 'real_esrgan_x2', 'real_esrgan_x2_fp16', 'real_esrgan_x4', 'real_esrgan_x4_fp16', 'real_esrgan_x8', 'real_esrgan_x8_fp16', 'real_hatgan_x4', 'real_web_photo_x4', 'realistic_rescaler_x4', 'remacri_x4', 'siax_x4', 'span_kendata_x4', 'swin2_sr_x4', 'tghq_face_x8', 'ultra_sharp_x4', 'ultra_sharp_2_x4' ]
+frame_enhancer_models : List[FrameEnhancerModel] = list(get_args(FrameEnhancerModel))
 
 frame_enhancer_blend_range : Sequence[int] = create_int_range(0, 100, 1)

facefusion/processors/modules/frame_enhancer/core.py

@@ -1,9 +1,12 @@
 from argparse import ArgumentParser
 from functools import lru_cache
+from types import ModuleType
+from typing import List
 
 import cv2
 import numpy
 
+import facefusion.choices
 import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, inference_manager, logger, state_manager, translator, video_manager
@@ -16,8 +19,8 @@ from facefusion.processors.modules.frame_enhancer.types import FrameEnhancerInpu
 from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import conditional_thread_semaphore
-from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
+from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, InferenceProvider, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import blend_frame, create_tile_frames, merge_tile_frames, read_static_image, read_static_video_frame
+from facefusion.vision import blend_frame, create_tile_frames, merge_tile_frames, read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
@@ -56,7 +59,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 			'__metadata__':
 			{
 				'vendor': 'Helaman',
-				'license': 'Non-Commercial',
+				'license': 'CC-BY-4.0',
 				'year': 2023
 			},
 			'hashes':
@@ -83,7 +86,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 			'__metadata__':
 			{
 				'vendor': 'Phhofm',
-				'license': 'Non-Commercial',
+				'license': 'CC-BY-4.0',
 				'year': 2023
 			},
 			'hashes':
@@ -156,6 +159,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/real_esrgan_x2_fp16.onnx')
 				}
 			},
+			'precision': 'fp16',
 			'size': (256, 16, 8),
 			'scale': 2
 		},
@@ -210,6 +214,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/real_esrgan_x4_fp16.onnx')
 				}
 			},
+			'precision': 'fp16',
 			'size': (256, 16, 8),
 			'scale': 4
 		},
@@ -264,6 +269,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 					'path': resolve_relative_path('../.assets/models/real_esrgan_x8_fp16.onnx')
 				}
 			},
+			'precision': 'fp16',
 			'size': (256, 16, 8),
 			'scale': 8
 		},
@@ -299,7 +305,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 			'__metadata__':
 			{
 				'vendor': 'Helaman',
-				'license': 'Non-Commercial',
+				'license': 'CC-BY-4.0',
 				'year': 2024
 			},
 			'hashes':
@@ -541,35 +547,38 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:
 
 
 def get_inference_pool() -> InferencePool:
-	model_names = [ get_frame_enhancer_model() ]
+	model_names = [ state_manager.get_item('frame_enhancer_model') ]
 	model_source_set = get_model_options().get('sources')
 
 	return inference_manager.get_inference_pool(__name__, model_names, model_source_set)
 
 
 def clear_inference_pool() -> None:
-	model_names = [ get_frame_enhancer_model() ]
+	model_names = [ state_manager.get_item('frame_enhancer_model') ]
 	inference_manager.clear_inference_pool(__name__, model_names)
 
 
+def resolve_inference_providers() -> List[InferenceProvider]:
+	model_precision = get_model_options().get('precision')
+
+	if is_macos() and has_execution_provider('coreml') and model_precision == 'fp16':
+		return\
+		[
+			(facefusion.choices.execution_provider_set.get('coreml'),
+			{
+				'ModelFormat': 'MLProgram',
+				'SpecializationStrategy': 'FastPrediction'
+			})
+		]
+
+	return []
+
+
 def get_model_options() -> ModelOptions:
-	model_name = get_frame_enhancer_model()
+	model_name = state_manager.get_item('frame_enhancer_model')
 	return create_static_model_set('full').get(model_name)
 
 
-def get_frame_enhancer_model() -> str:
-	frame_enhancer_model = state_manager.get_item('frame_enhancer_model')
-
-	if is_macos() and has_execution_provider('coreml'):
-		if frame_enhancer_model == 'real_esrgan_x2_fp16':
-			return 'real_esrgan_x2'
-		if frame_enhancer_model == 'real_esrgan_x4_fp16':
-			return 'real_esrgan_x4'
-		if frame_enhancer_model == 'real_esrgan_x8_fp16':
-			return 'real_esrgan_x8'
-	return frame_enhancer_model
-
-
 def register_args(program : ArgumentParser) -> None:
 	group_processors = find_argument_group(program, 'processors')
 	if group_processors:
@@ -583,10 +592,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('frame_enhancer_blend', args.get('frame_enhancer_blend'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -606,11 +623,15 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
+			common_module.clear_inference_pool()
 
 
 def enhance_frame(temp_vision_frame : VisionFrame) -> VisionFrame:

facefusion/processors/modules/frame_enhancer/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/frame_enhancer/types.py

@@ -1,10 +1,10 @@
-from typing import Literal, TypedDict
+from typing import List, Literal, TypedDict
 
 from facefusion.types import Mask, VisionFrame
 
 FrameEnhancerInputs = TypedDict('FrameEnhancerInputs',
 {
-	'target_vision_frame' : VisionFrame,
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/processors/modules/lip_syncer/choices.py

@@ -1,8 +1,8 @@
-from typing import List, Sequence
+from typing import List, Sequence, get_args
 
 from facefusion.common_helper import create_float_range
 from facefusion.processors.modules.lip_syncer.types import LipSyncerModel
 
-lip_syncer_models : List[LipSyncerModel] = [ 'edtalk_256', 'wav2lip_96', 'wav2lip_gan_96' ]
+lip_syncer_models : List[LipSyncerModel] = list(get_args(LipSyncerModel))
 
 lip_syncer_weight_range : Sequence[float] = create_float_range(0.0, 1.0, 0.05)

facefusion/processors/modules/lip_syncer/core.py

@@ -1,5 +1,7 @@
 from argparse import ArgumentParser
 from functools import lru_cache
+from types import ModuleType
+from typing import List
 
 import cv2
 import numpy
@@ -8,9 +10,9 @@ import facefusion.jobs.job_manager
 import facefusion.jobs.job_store
 from facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager, voice_extractor
 from facefusion.audio import read_static_voice
-from facefusion.common_helper import create_float_metavar
+from facefusion.common_helper import create_float_metavar, get_middle
 from facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url
-from facefusion.face_analyser import scale_face
+from facefusion.face_creator import scale_face
 from facefusion.face_helper import create_bounding_box, paste_back, warp_face_by_bounding_box, warp_face_by_face_landmark_5
 from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_mask
 from facefusion.face_selector import select_faces
@@ -21,7 +23,7 @@ from facefusion.processors.types import ProcessorOutputs
 from facefusion.program_helper import find_argument_group
 from facefusion.thread_helper import conditional_thread_semaphore
 from facefusion.types import ApplyStateItem, Args, AudioFrame, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame
-from facefusion.vision import read_static_image, read_static_video_frame
+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame
 
 
 @lru_cache()
@@ -142,10 +144,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('lip_syncer_weight', args.get('lip_syncer_weight'))
 
 
+def get_common_modules() -> List[ModuleType]:
+	return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, voice_extractor ]
+
+
 def pre_check() -> bool:
 	model_hash_set = get_model_options().get('hashes')
 	model_source_set = get_model_options().get('sources')
 
+	for common_module in get_common_modules():
+		if not common_module.pre_check():
+			return False
+
 	return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)
 
 
@@ -159,18 +169,16 @@ def pre_process(mode : ProcessMode) -> bool:
 def post_process() -> None:
 	read_static_image.cache_clear()
 	read_static_video_frame.cache_clear()
+	read_static_video_chunk.cache_clear()
 	read_static_voice.cache_clear()
 	video_manager.clear_video_pool()
+
 	if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:
 		clear_inference_pool()
+
 	if state_manager.get_item('video_memory_strategy') == 'strict':
-		content_analyser.clear_inference_pool()
+		for common_module in get_common_modules():
-		face_classifier.clear_inference_pool()
+			common_module.clear_inference_pool()
-		face_detector.clear_inference_pool()
-		face_landmarker.clear_inference_pool()
-		face_masker.clear_inference_pool()
-		face_recognizer.clear_inference_pool()
-		voice_extractor.clear_inference_pool()
 
 
 def sync_lip(target_face : Face, source_voice_frame : AudioFrame, temp_vision_frame : VisionFrame) -> VisionFrame:
@@ -282,11 +290,14 @@ def normalize_crop_frame(crop_vision_frame : VisionFrame) -> VisionFrame:
 
 def process_frame(inputs : LipSyncerInputs) -> ProcessorOutputs:
 	reference_vision_frame = inputs.get('reference_vision_frame')
+	source_vision_frames = inputs.get('source_vision_frames')
 	source_voice_frame = inputs.get('source_voice_frame')
-	target_vision_frame = inputs.get('target_vision_frame')
+	target_vision_frames = inputs.get('target_vision_frames')
 	temp_vision_frame = inputs.get('temp_vision_frame')
 	temp_vision_mask = inputs.get('temp_vision_mask')
-	target_faces = select_faces(reference_vision_frame, target_vision_frame)
+
+	target_vision_frame = get_middle(target_vision_frames)
+	target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)
 
 	if target_faces:
 		for target_face in target_faces:

facefusion/processors/modules/lip_syncer/locales.py

@@ -1,6 +1,6 @@
-from facefusion.types import Locals
+from facefusion.types import Locales
 
-LOCALS : Locals =\
+LOCALES : Locales =\
 {
 	'en':
 	{

facefusion/processors/modules/lip_syncer/types.py

@@ -1,4 +1,4 @@
-from typing import Any, Literal, TypeAlias, TypedDict
+from typing import Any, List, Literal, TypeAlias, TypedDict
 
 from numpy.typing import NDArray
 
@@ -7,8 +7,9 @@ from facefusion.types import AudioFrame, Mask, VisionFrame
 LipSyncerInputs = TypedDict('LipSyncerInputs',
 {
 	'reference_vision_frame' : VisionFrame,
+	'source_vision_frames' : List[VisionFrame],
 	'source_voice_frame' : AudioFrame,
-	'target_vision_frame' : VisionFrame,
+	'target_vision_frames' : List[VisionFrame],
 	'temp_vision_frame' : VisionFrame,
 	'temp_vision_mask' : Mask
 })

facefusion/program.py

@@ -10,7 +10,7 @@ from facefusion.ffmpeg import get_available_encoder_set
 from facefusion.filesystem import get_file_name, resolve_file_paths
 from facefusion.jobs import job_store
 from facefusion.processors.core import get_processors_modules
-from facefusion.sanitizer import sanitize_int_range
+from facefusion.sanitizer import sanitize_int_range, sanitize_job_id
 
 
 def create_help_formatter_small(prog : str) -> HelpFormatter:
@@ -133,6 +133,14 @@ def create_face_selector_program() -> ArgumentParser:
 	return program
 
 
+def create_face_tracker_program() -> ArgumentParser:
+	program = ArgumentParser(add_help = False)
+	group_face_tracker = program.add_argument_group('face tracker')
+	group_face_tracker.add_argument('--face-tracker-score', help = translator.get('help.face_tracker_score'), type = float, default = config.get_float_value('face_tracker', 'face_tracker_score', '0.0'), choices = facefusion.choices.face_tracker_score_range, metavar = create_float_metavar(facefusion.choices.face_tracker_score_range))
+	job_store.register_step_keys([ 'face_tracker_score' ])
+	return program
+
+
 def create_face_masker_program() -> ArgumentParser:
 	program = ArgumentParser(add_help = False)
 	group_face_masker = program.add_argument_group('face masker')
@@ -166,6 +174,14 @@ def create_frame_extraction_program() -> ArgumentParser:
 	return program
 
 
+def create_frame_distribution_program() -> ArgumentParser:
+	program = ArgumentParser(add_help = False)
+	group_frame_distribution = program.add_argument_group('frame distribution')
+	group_frame_distribution.add_argument('--target-frame-amount', help = translator.get('help.target_frame_amount'), type = int, default = config.get_int_value('frame_distribution', 'target_frame_amount', '2'), choices = facefusion.choices.target_frame_amount_range, metavar = create_int_metavar(facefusion.choices.target_frame_amount_range))
+	job_store.register_step_keys([ 'target_frame_amount' ])
+	return program
+
+
 def create_output_creation_program() -> ArgumentParser:
 	program = ArgumentParser(add_help = False)
 	available_encoder_set = get_available_encoder_set()
@@ -188,7 +204,7 @@ def create_processors_program() -> ArgumentParser:
 	program = ArgumentParser(add_help = False)
 	available_processors = [ get_file_name(file_path) for file_path in resolve_file_paths('facefusion/processors/modules') ]
 	group_processors = program.add_argument_group('processors')
-	group_processors.add_argument('--processors', help = translator.get('help.processors').format(choices = ', '.join(available_processors)), default = config.get_str_list('processors', 'processors', 'face_swapper'), nargs = '+')
+	group_processors.add_argument('--processors', help = translator.get('help.processors').format(choices = ', '.join(available_processors)), default = config.get_str_list('processors', 'processors', 'face_swapper'), choices = available_processors, nargs = '+', metavar = 'PROCESSORS')
 	job_store.register_step_keys([ 'processors' ])
 	for processor_module in get_processors_modules(available_processors):
 		processor_module.register_args(program)
@@ -200,7 +216,7 @@ def create_uis_program() -> ArgumentParser:
 	available_ui_layouts = [ get_file_name(file_path) for file_path in resolve_file_paths('facefusion/uis/layouts') ]
 	group_uis = program.add_argument_group('uis')
 	group_uis.add_argument('--open-browser', help = translator.get('help.open_browser'), action = 'store_true', default = config.get_bool_value('uis', 'open_browser'))
-	group_uis.add_argument('--ui-layouts', help = translator.get('help.ui_layouts').format(choices = ', '.join(available_ui_layouts)), default = config.get_str_list('uis', 'ui_layouts', 'default'), nargs = '+')
+	group_uis.add_argument('--ui-layouts', help = translator.get('help.ui_layouts').format(choices = ', '.join(available_ui_layouts)), default = config.get_str_list('uis', 'ui_layouts', 'default'), choices = available_ui_layouts, nargs = '+', metavar = 'UI_LAYOUTS')
 	group_uis.add_argument('--ui-workflow', help = translator.get('help.ui_workflow'), default = config.get_str_value('uis', 'ui_workflow', 'instant_runner'), choices = facefusion.choices.ui_workflows)
 	return program
 
@@ -245,8 +261,7 @@ def create_memory_program() -> ArgumentParser:
 	program = ArgumentParser(add_help = False)
 	group_memory = program.add_argument_group('memory')
 	group_memory.add_argument('--video-memory-strategy', help = translator.get('help.video_memory_strategy'), default = config.get_str_value('memory', 'video_memory_strategy', 'strict'), choices = facefusion.choices.video_memory_strategies)
-	group_memory.add_argument('--system-memory-limit', help = translator.get('help.system_memory_limit'), type = int, default = config.get_int_value('memory', 'system_memory_limit', '0'), choices = facefusion.choices.system_memory_limit_range, metavar = create_int_metavar(facefusion.choices.system_memory_limit_range))
+	job_store.register_job_keys([ 'video_memory_strategy' ])
-	job_store.register_job_keys([ 'video_memory_strategy', 'system_memory_limit' ])
 	return program
 
 
@@ -268,8 +283,7 @@ def create_halt_on_error_program() -> ArgumentParser:
 
 def create_job_id_program() -> ArgumentParser:
 	program = ArgumentParser(add_help = False)
-	program.add_argument('job_id', help = translator.get('help.job_id'))
+	program.add_argument('job_id', help = translator.get('help.job_id'), type = sanitize_job_id)
-	job_store.register_job_keys([ 'job_id' ])
 	return program
 
 
@@ -286,7 +300,7 @@ def create_step_index_program() -> ArgumentParser:
 
 
 def collect_step_program() -> ArgumentParser:
-	return ArgumentParser(parents = [ create_face_detector_program(), create_face_landmarker_program(), create_face_selector_program(), create_face_masker_program(), create_voice_extractor_program(), create_frame_extraction_program(), create_output_creation_program(), create_processors_program() ], add_help = False)
+	return ArgumentParser(parents = [ create_face_detector_program(), create_face_landmarker_program(), create_face_selector_program(), create_face_tracker_program(), create_face_masker_program(), create_voice_extractor_program(), create_frame_extraction_program(), create_frame_distribution_program(), create_output_creation_program(), create_processors_program() ], add_help = False)
 
 
 def collect_job_program() -> ArgumentParser:
@@ -298,13 +312,11 @@ def create_program() -> ArgumentParser:
 	program._positionals.title = 'commands'
 	program.add_argument('-v', '--version', version = metadata.get('name') + ' ' + metadata.get('version'), action = 'version')
 	sub_program = program.add_subparsers(dest = 'command')
-	# general
 	sub_program.add_parser('run', help = translator.get('help.run'), parents = [ create_config_path_program(), create_temp_path_program(), create_jobs_path_program(), create_source_paths_program(), create_target_path_program(), create_output_path_program(), collect_step_program(), create_uis_program(), create_benchmark_program(), collect_job_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('headless-run', help = translator.get('help.headless_run'), parents = [ create_config_path_program(), create_temp_path_program(), create_jobs_path_program(), create_source_paths_program(), create_target_path_program(), create_output_path_program(), collect_step_program(), collect_job_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('batch-run', help = translator.get('help.batch_run'), parents = [ create_config_path_program(), create_temp_path_program(), create_jobs_path_program(), create_source_pattern_program(), create_target_pattern_program(), create_output_pattern_program(), collect_step_program(), collect_job_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('force-download', help = translator.get('help.force_download'), parents = [ create_download_providers_program(), create_download_scope_program(), create_log_level_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('benchmark', help = translator.get('help.benchmark'), parents = [ create_temp_path_program(), collect_step_program(), create_benchmark_program(), collect_job_program() ], formatter_class = create_help_formatter_large)
-	# job manager
 	sub_program.add_parser('job-list', help = translator.get('help.job_list'), parents = [ create_job_status_program(), create_jobs_path_program(), create_log_level_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('job-create', help = translator.get('help.job_create'), parents = [ create_job_id_program(), create_jobs_path_program(), create_log_level_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('job-submit', help = translator.get('help.job_submit'), parents = [ create_job_id_program(), create_jobs_path_program(), create_log_level_program() ], formatter_class = create_help_formatter_large)
@@ -315,7 +327,6 @@ def create_program() -> ArgumentParser:
 	sub_program.add_parser('job-remix-step', help = translator.get('help.job_remix_step'), parents = [ create_job_id_program(), create_step_index_program(), create_config_path_program(), create_jobs_path_program(), create_source_paths_program(), create_output_path_program(), collect_step_program(), create_log_level_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('job-insert-step', help = translator.get('help.job_insert_step'), parents = [ create_job_id_program(), create_step_index_program(), create_config_path_program(), create_jobs_path_program(), create_source_paths_program(), create_target_path_program(), create_output_path_program(), collect_step_program(), create_log_level_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('job-remove-step', help = translator.get('help.job_remove_step'), parents = [ create_job_id_program(), create_step_index_program(), create_jobs_path_program(), create_log_level_program() ], formatter_class = create_help_formatter_large)
-	# job runner
 	sub_program.add_parser('job-run', help = translator.get('help.job_run'), parents = [ create_job_id_program(), create_config_path_program(), create_temp_path_program(), create_jobs_path_program(), collect_job_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('job-run-all', help = translator.get('help.job_run_all'), parents = [ create_config_path_program(), create_temp_path_program(), create_jobs_path_program(), collect_job_program(), create_halt_on_error_program() ], formatter_class = create_help_formatter_large)
 	sub_program.add_parser('job-retry', help = translator.get('help.job_retry'), parents = [ create_job_id_program(), create_config_path_program(), create_temp_path_program(), create_jobs_path_program(), collect_job_program() ], formatter_class = create_help_formatter_large)

facefusion/sanitizer.py

@@ -1,7 +1,21 @@
-from typing import Sequence
+import hashlib
+from typing import Any, Sequence
+
+from facefusion.common_helper import cast_int
 
 
-def sanitize_int_range(value : int, int_range : Sequence[int]) -> int:
+def sanitize_job_id(job_id : str) -> str:
+	__job_id__ = job_id.replace('-', '')
+
+	if __job_id__.isalnum():
+		return job_id
+
+	return hashlib.sha1(job_id.encode()).hexdigest()
+
+
+def sanitize_int_range(value : Any, int_range : Sequence[int]) -> int:
+	value = cast_int(value)
+
 	if value in int_range:
 		return value
 	return int_range[0]

facefusion/streamer.py

@@ -2,7 +2,7 @@ import os
 import subprocess
 from collections import deque
 from concurrent.futures import ThreadPoolExecutor
-from typing import Deque, Iterator
+from typing import Deque, Iterator, List
 
 import cv2
 import numpy
@@ -20,23 +20,24 @@ from facefusion.vision import extract_vision_mask, read_static_images
 
 def multi_process_capture(camera_capture : cv2.VideoCapture, camera_fps : Fps) -> Iterator[VisionFrame]:
 	capture_deque : Deque[VisionFrame] = deque()
+	source_vision_frames = read_static_images(state_manager.get_item('source_paths'))
 
 	with tqdm(desc = translator.get('streaming'), unit = 'frame', disable = state_manager.get_item('log_level') in [ 'warn', 'error' ]) as progress:
 		with ThreadPoolExecutor(max_workers = state_manager.get_item('execution_thread_count')) as executor:
 			futures = []
 
 			while camera_capture and camera_capture.isOpened():
-				_, capture_frame = camera_capture.read()
+				_, capture_vision_frame = camera_capture.read()
-				if analyse_stream(capture_frame, camera_fps):
+				if analyse_stream(capture_vision_frame, camera_fps):
 					camera_capture.release()
 
-				if numpy.any(capture_frame):
+				if numpy.any(capture_vision_frame):
-					future = executor.submit(process_stream_frame, capture_frame)
+					future = executor.submit(process_stream_frame, source_vision_frames, capture_vision_frame)
 					futures.append(future)
 
 				for future_done in [ future for future in futures if future.done() ]:
-					capture_frame = future_done.result()
+					capture_vision_frame = future_done.result()
-					capture_deque.append(capture_frame)
+					capture_deque.append(capture_vision_frame)
 					futures.remove(future_done)
 
 				while capture_deque:
@@ -44,8 +45,7 @@ def multi_process_capture(camera_capture : cv2.VideoCapture, camera_fps : Fps) -
 					yield capture_deque.popleft()
 
 
-def process_stream_frame(target_vision_frame : VisionFrame) -> VisionFrame:
+def process_stream_frame(source_vision_frames : List[VisionFrame], target_vision_frame : VisionFrame) -> VisionFrame:
-	source_vision_frames = read_static_images(state_manager.get_item('source_paths'))
 	source_audio_frame = create_empty_audio_frame()
 	source_voice_frame = create_empty_audio_frame()
 	temp_vision_frame = target_vision_frame.copy()
@@ -60,7 +60,7 @@ def process_stream_frame(target_vision_frame : VisionFrame) -> VisionFrame:
 				'source_vision_frames': source_vision_frames,
 				'source_audio_frame': source_audio_frame,
 				'source_voice_frame': source_voice_frame,
-				'target_vision_frame': target_vision_frame,
+				'target_vision_frames': [ target_vision_frame ],
 				'temp_vision_frame': temp_vision_frame,
 				'temp_vision_mask': temp_vision_mask
 			})

facefusion/temp_helper.py

@@ -1,8 +1,8 @@
 import os
-from typing import List
 
 from facefusion import state_manager
 from facefusion.filesystem import create_directory, get_file_extension, get_file_name, move_file, remove_directory, resolve_file_pattern
+from facefusion.types import FrameSet
 
 
 def get_temp_file_path(file_path : str) -> str:
@@ -16,12 +16,18 @@ def move_temp_file(file_path : str, move_path : str) -> bool:
 	return move_file(temp_file_path, move_path)
 
 
-def resolve_temp_frame_paths(target_path : str) -> List[str]:
+def resolve_temp_frame_set(target_path : str) -> FrameSet:
-	temp_frames_pattern = get_temp_frames_pattern(target_path, '*')
+	temp_frame_pattern = get_temp_frame_pattern(target_path, '*')
-	return resolve_file_pattern(temp_frames_pattern)
+	temp_frame_set = {}
+
+	for temp_frame_path in resolve_file_pattern(temp_frame_pattern):
+		frame_number = int(get_file_name(temp_frame_path))
+		temp_frame_set[frame_number] = temp_frame_path
+
+	return temp_frame_set
 
 
-def get_temp_frames_pattern(target_path : str, temp_frame_prefix : str) -> str:
+def get_temp_frame_pattern(target_path : str, temp_frame_prefix : str) -> str:
 	temp_directory_path = get_temp_directory_path(target_path)
 	return os.path.join(temp_directory_path, temp_frame_prefix + '.' + state_manager.get_item('temp_frame_format'))
 

facefusion/translator.py

@@ -1,29 +1,29 @@
 import importlib
 from typing import Optional
 
-from facefusion.types import Language, LocalPoolSet, Locals
+from facefusion.types import Language, LocalePoolSet, Locales
 
-LOCAL_POOL_SET : LocalPoolSet = {}
+LOCALE_POOL_SET : LocalePoolSet = {}
 CURRENT_LANGUAGE : Language = 'en'
 
 
 def __autoload__(module_name : str) -> None:
 	try:
-		__locals__ = importlib.import_module(module_name + '.locals')
+		__locales__ = importlib.import_module(module_name + '.locales')
-		load(__locals__.LOCALS, module_name)
+		load(__locales__.LOCALES, module_name)
 	except ImportError:
 		pass
 
 
-def load(__locals__ : Locals, module_name : str) -> None:
+def load(__locales__ : Locales, module_name : str) -> None:
-	LOCAL_POOL_SET[module_name] = __locals__
+	LOCALE_POOL_SET[module_name] = __locales__
 
 
 def get(notation : str, module_name : str = 'facefusion') -> Optional[str]:
-	if module_name not in LOCAL_POOL_SET:
+	if module_name not in LOCALE_POOL_SET:
 		__autoload__(module_name)
 
-	current = LOCAL_POOL_SET.get(module_name).get(CURRENT_LANGUAGE)
+	current = LOCALE_POOL_SET.get(module_name).get(CURRENT_LANGUAGE)
 
 	for fragment in notation.split('.'):
 		if fragment in current:

facefusion/types.py

@@ -1,5 +1,6 @@
 from collections import namedtuple
-from typing import Any, Callable, Dict, List, Literal, Optional, Tuple, TypeAlias, TypedDict
+from threading import Lock
+from typing import Any, Callable, Dict, List, Literal, NotRequired, Optional, Tuple, TypeAlias, TypedDict
 
 import cv2
 import numpy
@@ -29,42 +30,50 @@ FaceScoreSet = TypedDict('FaceScoreSet',
 	'landmarker' : Score
 })
 Embedding : TypeAlias = NDArray[numpy.float64]
-Gender = Literal['female', 'male']
+
 Age : TypeAlias = range
+Gender = Literal['female', 'male']
 Race = Literal['white', 'black', 'latino', 'asian', 'indian', 'arabic']
+
+FaceSelectorGender = Literal['auto', 'female', 'male']
+FaceSelectorRace = Literal['auto', 'white', 'black', 'latino', 'asian', 'indian', 'arabic']
+
 Face = namedtuple('Face',
 [
+	'origin',
 	'bounding_box',
 	'score_set',
 	'landmark_set',
 	'angle',
 	'embedding',
 	'embedding_norm',
-	'gender',
 	'age',
+	'gender',
 	'race'
 ])
-FaceSet : TypeAlias = Dict[str, List[Face]]
+FaceSet = TypedDict('FaceSet',
-FaceStore = TypedDict('FaceStore',
 {
-	'static_faces' : FaceSet
+	'lock': Lock,
+	'faces': NotRequired[List[Face]]
 })
+FaceStore : TypeAlias = Dict[str, FaceSet]
+FaceTrack : TypeAlias = Dict[int, Face]
 
 Language = Literal['en']
-Locals : TypeAlias = Dict[Language, Dict[str, Any]]
+Locales : TypeAlias = Dict[Language, Dict[str, Any]]
-LocalPoolSet : TypeAlias = Dict[str, Locals]
+LocalePoolSet : TypeAlias = Dict[str, Locales]
 
 VideoCaptureSet : TypeAlias = Dict[str, cv2.VideoCapture]
 VideoWriterSet : TypeAlias = Dict[str, cv2.VideoWriter]
 CameraCaptureSet : TypeAlias = Dict[str, cv2.VideoCapture]
 VideoPoolSet = TypedDict('VideoPoolSet',
 {
-	'capture': VideoCaptureSet,
+	'capture' : VideoCaptureSet,
-	'writer': VideoWriterSet
+	'writer' : VideoWriterSet
 })
 CameraPoolSet = TypedDict('CameraPoolSet',
 {
-	'capture': CameraCaptureSet
+	'capture' : CameraCaptureSet
 })
 
 ColorMode = Literal['rgb', 'rgba']
@@ -138,6 +147,8 @@ AudioTypeSet : TypeAlias = Dict[AudioFormat, str]
 ImageTypeSet : TypeAlias = Dict[ImageFormat, str]
 VideoTypeSet : TypeAlias = Dict[VideoFormat, str]
 
+FrameSet : TypeAlias = Dict[int, str]
+
 AudioEncoder = Literal['flac', 'aac', 'libmp3lame', 'libopus', 'libvorbis', 'pcm_s16le', 'pcm_s32le']
 VideoEncoder = Literal['libx264', 'libx264rgb', 'libx265', 'libvpx-vp9', 'h264_nvenc', 'hevc_nvenc', 'h264_amf', 'hevc_amf', 'h264_qsv', 'hevc_qsv', 'h264_videotoolbox', 'hevc_videotoolbox', 'rawvideo']
 EncoderSet = TypedDict('EncoderSet',
@@ -167,10 +178,11 @@ ModelOptions : TypeAlias = Dict[str, Any]
 ModelSet : TypeAlias = Dict[str, ModelOptions]
 ModelInitializer : TypeAlias = NDArray[Any]
 
-ExecutionProvider = Literal['cpu', 'coreml', 'cuda', 'directml', 'openvino', 'migraphx', 'rocm', 'tensorrt']
+ExecutionProvider = Literal['cuda', 'tensorrt', 'rocm', 'migraphx', 'coreml', 'openvino', 'qnn', 'directml', 'cpu']
-ExecutionProviderValue = Literal['CPUExecutionProvider', 'CoreMLExecutionProvider', 'CUDAExecutionProvider', 'DmlExecutionProvider', 'OpenVINOExecutionProvider', 'MIGraphXExecutionProvider', 'ROCMExecutionProvider', 'TensorrtExecutionProvider']
+ExecutionProviderValue = Literal['CPUExecutionProvider', 'CoreMLExecutionProvider', 'CUDAExecutionProvider', 'DmlExecutionProvider', 'OpenVINOExecutionProvider', 'MIGraphXExecutionProvider', 'QNNExecutionProvider', 'ROCMExecutionProvider', 'TensorrtExecutionProvider']
 ExecutionProviderSet : TypeAlias = Dict[ExecutionProvider, ExecutionProviderValue]
-InferenceSessionProvider : TypeAlias = Any
+InferenceProvider : TypeAlias = Any
+InferenceOptionSet : TypeAlias = Dict[str, Any]
 ValueAndUnit = TypedDict('ValueAndUnit',
 {
 	'value' : int,
@@ -289,6 +301,7 @@ StateKey = Literal\
 	'reference_face_position',
 	'reference_face_distance',
 	'reference_frame_number',
+	'face_tracker_score',
 	'face_occluder_model',
 	'face_parser_model',
 	'face_mask_types',
@@ -301,6 +314,7 @@ StateKey = Literal\
 	'trim_frame_end',
 	'temp_frame_format',
 	'keep_temp',
+	'target_frame_amount',
 	'output_image_quality',
 	'output_image_scale',
 	'output_audio_encoder',
@@ -319,7 +333,6 @@ StateKey = Literal\
 	'execution_providers',
 	'execution_thread_count',
 	'video_memory_strategy',
-	'system_memory_limit',
 	'log_level',
 	'halt_on_error',
 	'job_id',
@@ -345,20 +358,21 @@ State = TypedDict('State',
 	'benchmark_cycle_count' : int,
 	'face_detector_model' : FaceDetectorModel,
 	'face_detector_size' : str,
-	'face_detector_margin': Margin,
+	'face_detector_margin' : Margin,
 	'face_detector_angles' : List[Angle],
 	'face_detector_score' : Score,
 	'face_landmarker_model' : FaceLandmarkerModel,
 	'face_landmarker_score' : Score,
 	'face_selector_mode' : FaceSelectorMode,
 	'face_selector_order' : FaceSelectorOrder,
-	'face_selector_race' : Race,
+	'face_selector_race' : FaceSelectorRace,
-	'face_selector_gender' : Gender,
+	'face_selector_gender' : FaceSelectorGender,
 	'face_selector_age_start' : int,
 	'face_selector_age_end' : int,
 	'reference_face_position' : int,
 	'reference_face_distance' : float,
 	'reference_frame_number' : int,
+	'face_tracker_score' : Score,
 	'face_occluder_model' : FaceOccluderModel,
 	'face_parser_model' : FaceParserModel,
 	'face_mask_types' : List[FaceMaskType],
@@ -366,11 +380,12 @@ State = TypedDict('State',
 	'face_mask_regions' : List[FaceMaskRegion],
 	'face_mask_blur' : float,
 	'face_mask_padding' : Padding,
-	'voice_extractor_model': VoiceExtractorModel,
+	'voice_extractor_model' : VoiceExtractorModel,
 	'trim_frame_start' : int,
 	'trim_frame_end' : int,
 	'temp_frame_format' : TempFrameFormat,
 	'keep_temp' : bool,
+	'target_frame_amount' : int,
 	'output_image_quality' : int,
 	'output_image_scale' : Scale,
 	'output_audio_encoder' : AudioEncoder,
@@ -389,7 +404,6 @@ State = TypedDict('State',
 	'execution_providers' : List[ExecutionProvider],
 	'execution_thread_count' : int,
 	'video_memory_strategy' : VideoMemoryStrategy,
-	'system_memory_limit' : int,
 	'log_level' : LogLevel,
 	'halt_on_error' : bool,
 	'job_id' : str,

facefusion/uis/assets/overrides.css

@@ -62,10 +62,14 @@
 	width: 1.125rem;
 }
 
-:root:root:root:root .thumbnail-item
+:root:root:root:root .gallery-container .thumbnail-item
 {
 	border: unset;
-	box-shadow: unset;
+}
+
+:root:root:root:root .gallery-container .grid-container
+{
+	grid-template-columns: repeat(7, 1fr);
 }
 
 :root:root:root:root .grid-wrap.fixed-height

facefusion/uis/choices.py

@@ -14,7 +14,7 @@ preview_resolutions : List[str] = [ '512x512', '768x768', '1024x1024' ]
 webcam_modes : List[WebcamMode] = [ 'inline', 'udp', 'v4l2' ]
 webcam_resolutions : List[str] = [ '320x240', '640x480', '800x600', '1024x768', '1280x720', '1280x960', '1920x1080' ]
 
-background_remover_colors : Dict[str, Color] =\
+background_remover_fill_colors : Dict[str, Color] =\
 {
 	'red' : (255, 0, 0, 255),
 	'green' : (0, 255, 0, 255),

facefusion/uis/components/background_remover_options.py

@@ -11,82 +11,132 @@ from facefusion.sanitizer import sanitize_int_range
 from facefusion.uis.core import get_ui_component, register_ui_component
 
 BACKGROUND_REMOVER_MODEL_DROPDOWN : Optional[gradio.Dropdown] = None
-BACKGROUND_REMOVER_COLOR_WRAPPER : Optional[gradio.Group] = None
+BACKGROUND_REMOVER_FILL_COLOR_WRAPPER : Optional[gradio.Group] = None
-BACKGROUND_REMOVER_COLOR_RED_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_FILL_COLOR_RED_NUMBER : Optional[gradio.Number] = None
-BACKGROUND_REMOVER_COLOR_GREEN_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_FILL_COLOR_GREEN_NUMBER : Optional[gradio.Number] = None
-BACKGROUND_REMOVER_COLOR_BLUE_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_FILL_COLOR_BLUE_NUMBER : Optional[gradio.Number] = None
-BACKGROUND_REMOVER_COLOR_ALPHA_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_FILL_COLOR_ALPHA_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_DESPILL_COLOR_WRAPPER : Optional[gradio.Group] = None
+BACKGROUND_REMOVER_DESPILL_COLOR_RED_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_DESPILL_COLOR_GREEN_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_DESPILL_COLOR_BLUE_NUMBER : Optional[gradio.Number] = None
+BACKGROUND_REMOVER_DESPILL_COLOR_ALPHA_NUMBER : Optional[gradio.Number] = None
 
 
 def render() -> None:
 	global BACKGROUND_REMOVER_MODEL_DROPDOWN
-	global BACKGROUND_REMOVER_COLOR_WRAPPER
+	global BACKGROUND_REMOVER_FILL_COLOR_WRAPPER
-	global BACKGROUND_REMOVER_COLOR_RED_NUMBER
+	global BACKGROUND_REMOVER_FILL_COLOR_RED_NUMBER
-	global BACKGROUND_REMOVER_COLOR_GREEN_NUMBER
+	global BACKGROUND_REMOVER_FILL_COLOR_GREEN_NUMBER
-	global BACKGROUND_REMOVER_COLOR_BLUE_NUMBER
+	global BACKGROUND_REMOVER_FILL_COLOR_BLUE_NUMBER
-	global BACKGROUND_REMOVER_COLOR_ALPHA_NUMBER
+	global BACKGROUND_REMOVER_FILL_COLOR_ALPHA_NUMBER
+	global BACKGROUND_REMOVER_DESPILL_COLOR_WRAPPER
+	global BACKGROUND_REMOVER_DESPILL_COLOR_RED_NUMBER
+	global BACKGROUND_REMOVER_DESPILL_COLOR_GREEN_NUMBER
+	global BACKGROUND_REMOVER_DESPILL_COLOR_BLUE_NUMBER
+	global BACKGROUND_REMOVER_DESPILL_COLOR_ALPHA_NUMBER
 
 	has_background_remover = 'background_remover' in state_manager.get_item('processors')
-	background_remover_color = state_manager.get_item('background_remover_color')
+	background_remover_fill_color = state_manager.get_item('background_remover_fill_color')
+	background_remover_despill_color = state_manager.get_item('background_remover_despill_color')
 	BACKGROUND_REMOVER_MODEL_DROPDOWN = gradio.Dropdown(
 		label = translator.get('uis.model_dropdown', 'facefusion.processors.modules.background_remover'),
 		choices = background_remover_choices.background_remover_models,
 		value = state_manager.get_item('background_remover_model'),
 		visible = has_background_remover
 	)
-	with gradio.Group(visible = has_background_remover) as BACKGROUND_REMOVER_COLOR_WRAPPER:
+	with gradio.Group(visible = has_background_remover) as BACKGROUND_REMOVER_FILL_COLOR_WRAPPER:
 		with gradio.Row():
-			BACKGROUND_REMOVER_COLOR_RED_NUMBER = gradio.Number(
+			BACKGROUND_REMOVER_FILL_COLOR_RED_NUMBER = gradio.Number(
-				label = translator.get('uis.color_red_number', 'facefusion.processors.modules.background_remover'),
+				label = translator.get('uis.fill_color_red_number', 'facefusion.processors.modules.background_remover'),
-				value = background_remover_color[0],
+				value = background_remover_fill_color[0],
 				minimum = background_remover_choices.background_remover_color_range[0],
 				maximum = background_remover_choices.background_remover_color_range[-1],
 				step = calculate_int_step(background_remover_choices.background_remover_color_range)
 			)
-			BACKGROUND_REMOVER_COLOR_GREEN_NUMBER = gradio.Number(
+			BACKGROUND_REMOVER_FILL_COLOR_GREEN_NUMBER = gradio.Number(
-				label = translator.get('uis.color_green_number', 'facefusion.processors.modules.background_remover'),
+				label = translator.get('uis.fill_color_green_number', 'facefusion.processors.modules.background_remover'),
-				value = background_remover_color[1],
+				value = background_remover_fill_color[1],
 				minimum = background_remover_choices.background_remover_color_range[0],
 				maximum = background_remover_choices.background_remover_color_range[-1],
 				step = calculate_int_step(background_remover_choices.background_remover_color_range)
 			)
 		with gradio.Row():
-			BACKGROUND_REMOVER_COLOR_BLUE_NUMBER = gradio.Number(
+			BACKGROUND_REMOVER_FILL_COLOR_BLUE_NUMBER = gradio.Number(
-				label = translator.get('uis.color_blue_number', 'facefusion.processors.modules.background_remover'),
+				label = translator.get('uis.fill_color_blue_number', 'facefusion.processors.modules.background_remover'),
-				value = background_remover_color[2],
+				value = background_remover_fill_color[2],
 				minimum = background_remover_choices.background_remover_color_range[0],
 				maximum = background_remover_choices.background_remover_color_range[-1],
 				step = calculate_int_step(background_remover_choices.background_remover_color_range)
 			)
-			BACKGROUND_REMOVER_COLOR_ALPHA_NUMBER = gradio.Number(
+			BACKGROUND_REMOVER_FILL_COLOR_ALPHA_NUMBER = gradio.Number(
-				label = translator.get('uis.color_alpha_number', 'facefusion.processors.modules.background_remover'),
+				label = translator.get('uis.fill_color_alpha_number', 'facefusion.processors.modules.background_remover'),
-				value = background_remover_color[3],
+				value = background_remover_fill_color[3],
+				minimum = background_remover_choices.background_remover_color_range[0],
+				maximum = background_remover_choices.background_remover_color_range[-1],
+				step = calculate_int_step(background_remover_choices.background_remover_color_range)
+			)
+	with gradio.Group(visible = has_background_remover) as BACKGROUND_REMOVER_DESPILL_COLOR_WRAPPER:
+		with gradio.Row():
+			BACKGROUND_REMOVER_DESPILL_COLOR_RED_NUMBER = gradio.Number(
+				label = translator.get('uis.despill_color_red_number', 'facefusion.processors.modules.background_remover'),
+				value = background_remover_despill_color[0],
+				minimum = background_remover_choices.background_remover_color_range[0],
+				maximum = background_remover_choices.background_remover_color_range[-1],
+				step = calculate_int_step(background_remover_choices.background_remover_color_range)
+			)
+			BACKGROUND_REMOVER_DESPILL_COLOR_GREEN_NUMBER = gradio.Number(
+				label = translator.get('uis.despill_color_green_number', 'facefusion.processors.modules.background_remover'),
+				value = background_remover_despill_color[1],
+				minimum = background_remover_choices.background_remover_color_range[0],
+				maximum = background_remover_choices.background_remover_color_range[-1],
+				step = calculate_int_step(background_remover_choices.background_remover_color_range)
+			)
+		with gradio.Row():
+			BACKGROUND_REMOVER_DESPILL_COLOR_BLUE_NUMBER = gradio.Number(
+				label = translator.get('uis.despill_color_blue_number', 'facefusion.processors.modules.background_remover'),
+				value = background_remover_despill_color[2],
+				minimum = background_remover_choices.background_remover_color_range[0],
+				maximum = background_remover_choices.background_remover_color_range[-1],
+				step = calculate_int_step(background_remover_choices.background_remover_color_range)
+			)
+			BACKGROUND_REMOVER_DESPILL_COLOR_ALPHA_NUMBER = gradio.Number(
+				label = translator.get('uis.despill_color_alpha_number', 'facefusion.processors.modules.background_remover'),
+				value = background_remover_despill_color[3],
 				minimum = background_remover_choices.background_remover_color_range[0],
 				maximum = background_remover_choices.background_remover_color_range[-1],
 				step = calculate_int_step(background_remover_choices.background_remover_color_range)
 			)
 	register_ui_component('background_remover_model_dropdown', BACKGROUND_REMOVER_MODEL_DROPDOWN)
-	register_ui_component('background_remover_color_red_number', BACKGROUND_REMOVER_COLOR_RED_NUMBER)
+	register_ui_component('background_remover_fill_color_red_number', BACKGROUND_REMOVER_FILL_COLOR_RED_NUMBER)
-	register_ui_component('background_remover_color_green_number', BACKGROUND_REMOVER_COLOR_GREEN_NUMBER)
+	register_ui_component('background_remover_fill_color_green_number', BACKGROUND_REMOVER_FILL_COLOR_GREEN_NUMBER)
-	register_ui_component('background_remover_color_blue_number', BACKGROUND_REMOVER_COLOR_BLUE_NUMBER)
+	register_ui_component('background_remover_fill_color_blue_number', BACKGROUND_REMOVER_FILL_COLOR_BLUE_NUMBER)
-	register_ui_component('background_remover_color_alpha_number', BACKGROUND_REMOVER_COLOR_ALPHA_NUMBER)
+	register_ui_component('background_remover_fill_color_alpha_number', BACKGROUND_REMOVER_FILL_COLOR_ALPHA_NUMBER)
+	register_ui_component('background_remover_despill_color_red_number', BACKGROUND_REMOVER_DESPILL_COLOR_RED_NUMBER)
+	register_ui_component('background_remover_despill_color_green_number', BACKGROUND_REMOVER_DESPILL_COLOR_GREEN_NUMBER)
+	register_ui_component('background_remover_despill_color_blue_number', BACKGROUND_REMOVER_DESPILL_COLOR_BLUE_NUMBER)
+	register_ui_component('background_remover_despill_color_alpha_number', BACKGROUND_REMOVER_DESPILL_COLOR_ALPHA_NUMBER)
 
 
 def listen() -> None:
 	BACKGROUND_REMOVER_MODEL_DROPDOWN.change(update_background_remover_model, inputs = BACKGROUND_REMOVER_MODEL_DROPDOWN, outputs = BACKGROUND_REMOVER_MODEL_DROPDOWN)
-	background_remover_color_inputs = [ BACKGROUND_REMOVER_COLOR_RED_NUMBER, BACKGROUND_REMOVER_COLOR_GREEN_NUMBER, BACKGROUND_REMOVER_COLOR_BLUE_NUMBER, BACKGROUND_REMOVER_COLOR_ALPHA_NUMBER ]
+	background_remover_fill_color_inputs = [ BACKGROUND_REMOVER_FILL_COLOR_RED_NUMBER, BACKGROUND_REMOVER_FILL_COLOR_GREEN_NUMBER, BACKGROUND_REMOVER_FILL_COLOR_BLUE_NUMBER, BACKGROUND_REMOVER_FILL_COLOR_ALPHA_NUMBER ]
+	background_remover_despill_color_inputs = [ BACKGROUND_REMOVER_DESPILL_COLOR_RED_NUMBER, BACKGROUND_REMOVER_DESPILL_COLOR_GREEN_NUMBER, BACKGROUND_REMOVER_DESPILL_COLOR_BLUE_NUMBER, BACKGROUND_REMOVER_DESPILL_COLOR_ALPHA_NUMBER ]
 
-	for background_remover_color_input in background_remover_color_inputs:
+	for background_remover_fill_color_input in background_remover_fill_color_inputs:
-		background_remover_color_input.change(update_background_remover_color, inputs = background_remover_color_inputs)
+		background_remover_fill_color_input.change(update_background_remover_fill_color, inputs = background_remover_fill_color_inputs)
+
+	for background_remover_despill_color_input in background_remover_despill_color_inputs:
+		background_remover_despill_color_input.change(update_background_remover_despill_color, inputs = background_remover_despill_color_inputs)
 
 	processors_checkbox_group = get_ui_component('processors_checkbox_group')
 	if processors_checkbox_group:
-		processors_checkbox_group.change(remote_update, inputs = processors_checkbox_group, outputs = [BACKGROUND_REMOVER_MODEL_DROPDOWN, BACKGROUND_REMOVER_COLOR_WRAPPER])
+		processors_checkbox_group.change(remote_update, inputs = processors_checkbox_group, outputs = [ BACKGROUND_REMOVER_MODEL_DROPDOWN, BACKGROUND_REMOVER_FILL_COLOR_WRAPPER, BACKGROUND_REMOVER_DESPILL_COLOR_WRAPPER ])
 
 
-def remote_update(processors : List[str]) -> Tuple[gradio.Dropdown, gradio.Group]:
+def remote_update(processors : List[str]) -> Tuple[gradio.Dropdown, gradio.Group, gradio.Group]:
 	has_background_remover = 'background_remover' in processors
-	return gradio.Dropdown(visible = has_background_remover), gradio.Group(visible = has_background_remover)
+	return gradio.Dropdown(visible = has_background_remover), gradio.Group(visible = has_background_remover), gradio.Group(visible = has_background_remover)
 
 
 def update_background_remover_model(background_remover_model : BackgroundRemoverModel) -> gradio.Dropdown:
@@ -99,9 +149,17 @@ def update_background_remover_model(background_remover_model : BackgroundRemover
 	return gradio.Dropdown()
 
 
-def update_background_remover_color(red : int, green : int, blue : int, alpha : int) -> None:
+def update_background_remover_fill_color(red : int, green : int, blue : int, alpha : int) -> None:
 	red = sanitize_int_range(red, background_remover_choices.background_remover_color_range)
 	green = sanitize_int_range(green, background_remover_choices.background_remover_color_range)
 	blue = sanitize_int_range(blue, background_remover_choices.background_remover_color_range)
 	alpha = sanitize_int_range(alpha, background_remover_choices.background_remover_color_range)
-	state_manager.set_item('background_remover_color', (red, green, blue, alpha))
+	state_manager.set_item('background_remover_fill_color', (red, green, blue, alpha))
+
+
+def update_background_remover_despill_color(red : int, green : int, blue : int, alpha : int) -> None:
+	red = sanitize_int_range(red, background_remover_choices.background_remover_color_range)
+	green = sanitize_int_range(green, background_remover_choices.background_remover_color_range)
+	blue = sanitize_int_range(blue, background_remover_choices.background_remover_color_range)
+	alpha = sanitize_int_range(alpha, background_remover_choices.background_remover_color_range)
+	state_manager.set_item('background_remover_despill_color', (red, green, blue, alpha))

facefusion/uis/components/face_selector.py

@@ -7,15 +7,15 @@ from gradio_rangeslider import RangeSlider
 import facefusion.choices
 from facefusion import state_manager, translator
 from facefusion.common_helper import calculate_float_step, calculate_int_step
-from facefusion.face_analyser import get_many_faces
+from facefusion.face_creator import get_many_faces
 from facefusion.face_selector import sort_and_filter_faces
-from facefusion.face_store import clear_static_faces
+from facefusion.face_store import clear_faces
-from facefusion.filesystem import is_image, is_video
+from facefusion.filesystem import filter_image_paths, is_image, is_video
-from facefusion.types import FaceSelectorMode, FaceSelectorOrder, Gender, Race, VisionFrame
+from facefusion.types import FaceSelectorGender, FaceSelectorMode, FaceSelectorOrder, FaceSelectorRace, VisionFrame
 from facefusion.uis.core import get_ui_component, get_ui_components, register_ui_component
 from facefusion.uis.types import ComponentOptions
 from facefusion.uis.ui_helper import convert_str_none
-from facefusion.vision import fit_cover_frame, read_static_image, read_video_frame
+from facefusion.vision import fit_cover_frame, read_static_image, read_static_images, read_video_frame
 
 FACE_SELECTOR_MODE_DROPDOWN : Optional[gradio.Dropdown] = None
 FACE_SELECTOR_ORDER_DROPDOWN : Optional[gradio.Dropdown] = None
@@ -39,17 +39,18 @@ def render() -> None:
 	{
 		'label': translator.get('uis.reference_face_gallery'),
 		'object_fit': 'cover',
-		'columns': 7,
 		'allow_preview': False,
 		'elem_classes': 'box-face-selector',
 		'visible': 'reference' in state_manager.get_item('face_selector_mode')
 	}
+	source_vision_frames = read_static_images(filter_image_paths(state_manager.get_item('source_paths')))
+
 	if is_image(state_manager.get_item('target_path')):
 		target_vision_frame = read_static_image(state_manager.get_item('target_path'))
-		reference_face_gallery_options['value'] = extract_gallery_frames(target_vision_frame)
+		reference_face_gallery_options['value'] = extract_gallery_frames(source_vision_frames, target_vision_frame)
 	if is_video(state_manager.get_item('target_path')):
 		target_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))
-		reference_face_gallery_options['value'] = extract_gallery_frames(target_vision_frame)
+		reference_face_gallery_options['value'] = extract_gallery_frames(source_vision_frames, target_vision_frame)
 	FACE_SELECTOR_MODE_DROPDOWN = gradio.Dropdown(
 		label = translator.get('uis.face_selector_mode_dropdown'),
 		choices = facefusion.choices.face_selector_modes,
@@ -156,12 +157,12 @@ def update_face_selector_order(face_analyser_order : FaceSelectorOrder) -> gradi
 	return update_reference_position_gallery()
 
 
-def update_face_selector_gender(face_selector_gender : Gender) -> gradio.Gallery:
+def update_face_selector_gender(face_selector_gender : FaceSelectorGender) -> gradio.Gallery:
 	state_manager.set_item('face_selector_gender', convert_str_none(face_selector_gender))
 	return update_reference_position_gallery()
 
 
-def update_face_selector_race(face_selector_race : Race) -> gradio.Gallery:
+def update_face_selector_race(face_selector_race : FaceSelectorRace) -> gradio.Gallery:
 	state_manager.set_item('face_selector_race', convert_str_none(face_selector_race))
 	return update_reference_position_gallery()
 
@@ -194,30 +195,33 @@ def clear_reference_frame_number() -> None:
 
 
 def clear_and_update_reference_position_gallery() -> gradio.Gallery:
-	clear_static_faces()
+	clear_faces()
 	return update_reference_position_gallery()
 
 
 def update_reference_position_gallery(frame_number : int = 0) -> gradio.Gallery:
 	gallery_vision_frames = []
+	source_vision_frames = read_static_images(filter_image_paths(state_manager.get_item('source_paths')))
+
 	if is_image(state_manager.get_item('target_path')):
 		target_vision_frame = read_static_image(state_manager.get_item('target_path'))
-		gallery_vision_frames = extract_gallery_frames(target_vision_frame)
+		gallery_vision_frames = extract_gallery_frames(source_vision_frames, target_vision_frame)
 	if is_video(state_manager.get_item('target_path')):
 		target_vision_frame = read_video_frame(state_manager.get_item('target_path'), frame_number)
-		gallery_vision_frames = extract_gallery_frames(target_vision_frame)
+		gallery_vision_frames = extract_gallery_frames(source_vision_frames, target_vision_frame)
 	if gallery_vision_frames:
 		return gradio.Gallery(value = gallery_vision_frames)
 	return gradio.Gallery(value = None)
 
 
-def extract_gallery_frames(target_vision_frame : VisionFrame) -> List[VisionFrame]:
+def extract_gallery_frames(source_vision_frames : List[VisionFrame], target_vision_frame : VisionFrame) -> List[VisionFrame]:
 	gallery_vision_frames = []
-	faces = get_many_faces([ target_vision_frame ])
+	source_faces = get_many_faces(source_vision_frames)
-	faces = sort_and_filter_faces(faces)
+	target_faces = get_many_faces([ target_vision_frame ])
+	target_faces = sort_and_filter_faces(source_faces, target_faces)
 
-	for face in faces:
+	for target_face in target_faces:
-		start_x, start_y, end_x, end_y = map(int, face.bounding_box)
+		start_x, start_y, end_x, end_y = map(int, target_face.bounding_box)
 		padding_x = int((end_x - start_x) * 0.25)
 		padding_y = int((end_y - start_y) * 0.25)
 		start_x = max(0, start_x - padding_x)
@@ -228,4 +232,5 @@ def extract_gallery_frames(target_vision_frame : VisionFrame) -> List[VisionFram
 		crop_vision_frame = fit_cover_frame(crop_vision_frame, (128, 128))
 		crop_vision_frame = cv2.cvtColor(crop_vision_frame, cv2.COLOR_BGR2RGB)
 		gallery_vision_frames.append(crop_vision_frame)
+
 	return gallery_vision_frames

facefusion/uis/components/face_tracker.py

@@ -0,0 +1,32 @@
+from typing import Optional
+
+import gradio
+
+import facefusion.choices
+from facefusion import state_manager, translator
+from facefusion.common_helper import calculate_float_step
+from facefusion.types import Score
+from facefusion.uis.core import register_ui_component
+
+FACE_TRACKER_SCORE_SLIDER : Optional[gradio.Slider] = None
+
+
+def render() -> None:
+	global FACE_TRACKER_SCORE_SLIDER
+
+	FACE_TRACKER_SCORE_SLIDER = gradio.Slider(
+		label = translator.get('uis.face_tracker_score_slider'),
+		value = state_manager.get_item('face_tracker_score'),
+		step = calculate_float_step(facefusion.choices.face_tracker_score_range),
+		minimum = facefusion.choices.face_tracker_score_range[0],
+		maximum = facefusion.choices.face_tracker_score_range[-1]
+	)
+	register_ui_component('face_tracker_score_slider', FACE_TRACKER_SCORE_SLIDER)
+
+
+def listen() -> None:
+	FACE_TRACKER_SCORE_SLIDER.release(update_face_tracker_score, inputs = FACE_TRACKER_SCORE_SLIDER)
+
+
+def update_face_tracker_score(face_tracker_score : Score) -> None:
+	state_manager.set_item('face_tracker_score', face_tracker_score)

facefusion/uis/components/memory.py

@@ -4,39 +4,24 @@ import gradio
 
 import facefusion.choices
 from facefusion import state_manager, translator
-from facefusion.common_helper import calculate_int_step
 from facefusion.types import VideoMemoryStrategy
 
 VIDEO_MEMORY_STRATEGY_DROPDOWN : Optional[gradio.Dropdown] = None
-SYSTEM_MEMORY_LIMIT_SLIDER : Optional[gradio.Slider] = None
 
 
 def render() -> None:
 	global VIDEO_MEMORY_STRATEGY_DROPDOWN
-	global SYSTEM_MEMORY_LIMIT_SLIDER
 
 	VIDEO_MEMORY_STRATEGY_DROPDOWN = gradio.Dropdown(
 		label = translator.get('uis.video_memory_strategy_dropdown'),
 		choices = facefusion.choices.video_memory_strategies,
 		value = state_manager.get_item('video_memory_strategy')
 	)
-	SYSTEM_MEMORY_LIMIT_SLIDER = gradio.Slider(
-		label = translator.get('uis.system_memory_limit_slider'),
-		step = calculate_int_step(facefusion.choices.system_memory_limit_range),
-		minimum = facefusion.choices.system_memory_limit_range[0],
-		maximum = facefusion.choices.system_memory_limit_range[-1],
-		value = state_manager.get_item('system_memory_limit')
-	)
 
 
 def listen() -> None:
 	VIDEO_MEMORY_STRATEGY_DROPDOWN.change(update_video_memory_strategy, inputs = VIDEO_MEMORY_STRATEGY_DROPDOWN)
-	SYSTEM_MEMORY_LIMIT_SLIDER.release(update_system_memory_limit, inputs = SYSTEM_MEMORY_LIMIT_SLIDER)
 
 
 def update_video_memory_strategy(video_memory_strategy : VideoMemoryStrategy) -> None:
 	state_manager.set_item('video_memory_strategy', video_memory_strategy)
-
-
-def update_system_memory_limit(system_memory_limit : float) -> None:
-	state_manager.set_item('system_memory_limit', int(system_memory_limit))

facefusion/uis/components/output_options.py

@@ -132,7 +132,7 @@ def listen() -> None:
 		'target_video'
 	]):
 		for method in [ 'change', 'clear' ]:
-			getattr(ui_component, method)(remote_update, outputs = [OUTPUT_IMAGE_QUALITY_SLIDER, OUTPUT_IMAGE_SCALE_SLIDER, OUTPUT_AUDIO_ENCODER_DROPDOWN, OUTPUT_AUDIO_QUALITY_SLIDER, OUTPUT_AUDIO_VOLUME_SLIDER, OUTPUT_VIDEO_ENCODER_DROPDOWN, OUTPUT_VIDEO_PRESET_DROPDOWN, OUTPUT_VIDEO_QUALITY_SLIDER, OUTPUT_VIDEO_SCALE_SLIDER, OUTPUT_VIDEO_FPS_SLIDER])
+			getattr(ui_component, method)(remote_update, outputs = [ OUTPUT_IMAGE_QUALITY_SLIDER, OUTPUT_IMAGE_SCALE_SLIDER, OUTPUT_AUDIO_ENCODER_DROPDOWN, OUTPUT_AUDIO_QUALITY_SLIDER, OUTPUT_AUDIO_VOLUME_SLIDER, OUTPUT_VIDEO_ENCODER_DROPDOWN, OUTPUT_VIDEO_PRESET_DROPDOWN, OUTPUT_VIDEO_QUALITY_SLIDER, OUTPUT_VIDEO_SCALE_SLIDER, OUTPUT_VIDEO_FPS_SLIDER ])
 
 
 def remote_update() -> Tuple[gradio.Slider, gradio.Slider, gradio.Dropdown, gradio.Slider, gradio.Slider, gradio.Dropdown, gradio.Dropdown, gradio.Slider, gradio.Slider, gradio.Slider]:

facefusion/uis/components/preview.py

@@ -7,18 +7,18 @@ import numpy
 
 from facefusion import logger, process_manager, state_manager, translator
 from facefusion.audio import create_empty_audio_frame, get_voice_frame
-from facefusion.common_helper import get_first
+from facefusion.common_helper import get_first, get_middle
 from facefusion.content_analyser import analyse_frame
-from facefusion.face_analyser import get_one_face
+from facefusion.face_creator import get_one_face
 from facefusion.face_selector import select_faces
-from facefusion.face_store import clear_static_faces
+from facefusion.face_store import clear_faces
 from facefusion.filesystem import filter_audio_paths, is_image, is_video
 from facefusion.processors.core import get_processors_modules
 from facefusion.types import AudioFrame, Face, Mask, VisionFrame
 from facefusion.uis import choices as uis_choices
 from facefusion.uis.core import get_ui_component, get_ui_components, register_ui_component
 from facefusion.uis.types import ComponentOptions, PreviewMode
-from facefusion.vision import detect_frame_orientation, extract_vision_mask, fit_cover_frame, merge_vision_mask, obscure_frame, read_static_image, read_static_images, read_video_frame, restrict_frame, unpack_resolution
+from facefusion.vision import detect_frame_orientation, extract_vision_mask, fit_cover_frame, merge_vision_mask, obscure_frame, read_static_image, read_static_images, read_video_frame, restrict_frame, select_video_frames, unpack_resolution
 
 PREVIEW_IMAGE : Optional[gradio.Image] = None
 
@@ -36,7 +36,7 @@ def render() -> None:
 	source_audio_frame = create_empty_audio_frame()
 	source_voice_frame = create_empty_audio_frame()
 
-	if source_audio_path and state_manager.get_item('output_video_fps') and state_manager.get_item('reference_frame_number'):
+	if source_audio_path and state_manager.get_item('output_video_fps'):
 		temp_voice_frame = get_voice_frame(source_audio_path, state_manager.get_item('output_video_fps'), state_manager.get_item('reference_frame_number'))
 		if numpy.any(temp_voice_frame):
 			source_voice_frame = temp_voice_frame
@@ -44,14 +44,14 @@ def render() -> None:
 	if is_image(state_manager.get_item('target_path')):
 		target_vision_frame = read_static_image(state_manager.get_item('target_path'))
 		reference_vision_frame = read_static_image(state_manager.get_item('target_path'))
-		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frame, uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])
+		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, [ target_vision_frame ], uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])
 		preview_image_options['value'] = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGR2RGB)
 		preview_image_options['elem_classes'] = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ]
 
 	if is_video(state_manager.get_item('target_path')):
-		temp_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))
 		reference_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))
-		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, temp_vision_frame, uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])
+		target_vision_frames = select_video_frames(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'), state_manager.get_item('target_frame_amount'))
+		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frames, uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])
 		preview_image_options['value'] = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGR2RGB)
 		preview_image_options['elem_classes'] = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ]
 		preview_image_options['visible'] = True
@@ -90,10 +90,14 @@ def listen() -> None:
 
 	for ui_component in get_ui_components(
 	[
-		'background_remover_color_red_number',
+		'background_remover_fill_color_red_number',
-		'background_remover_color_green_number',
+		'background_remover_fill_color_green_number',
-		'background_remover_color_blue_number',
+		'background_remover_fill_color_blue_number',
-		'background_remover_color_alpha_number',
+		'background_remover_fill_color_alpha_number',
+		'background_remover_despill_color_red_number',
+		'background_remover_despill_color_green_number',
+		'background_remover_despill_color_blue_number',
+		'background_remover_despill_color_alpha_number',
 		'face_debugger_items_checkbox_group',
 		'frame_colorizer_size_dropdown',
 		'face_mask_types_checkbox_group',
@@ -130,6 +134,7 @@ def listen() -> None:
 		'lip_syncer_weight_slider',
 		'reference_face_distance_slider',
 		'face_selector_age_range_slider',
+		'face_tracker_score_slider',
 		'face_mask_blur_slider',
 		'face_mask_padding_top_slider',
 		'face_mask_padding_bottom_slider',
@@ -185,43 +190,42 @@ def update_preview_image(preview_mode : PreviewMode, preview_resolution : str, f
 	source_audio_frame = create_empty_audio_frame()
 	source_voice_frame = create_empty_audio_frame()
 
-	if source_audio_path and state_manager.get_item('output_video_fps') and state_manager.get_item('reference_frame_number'):
+	if source_audio_path and state_manager.get_item('output_video_fps'):
-		reference_audio_frame_number = state_manager.get_item('reference_frame_number')
+		audio_frame_number = frame_number
 		if state_manager.get_item('trim_frame_start'):
-			reference_audio_frame_number -= state_manager.get_item('trim_frame_start')
+			audio_frame_number -= state_manager.get_item('trim_frame_start')
-		temp_voice_frame = get_voice_frame(source_audio_path, state_manager.get_item('output_video_fps'), reference_audio_frame_number)
+		temp_voice_frame = get_voice_frame(source_audio_path, state_manager.get_item('output_video_fps'), audio_frame_number)
 		if numpy.any(temp_voice_frame):
 			source_voice_frame = temp_voice_frame
 
 	if is_image(state_manager.get_item('target_path')):
 		reference_vision_frame = read_static_image(state_manager.get_item('target_path'))
 		target_vision_frame = read_static_image(state_manager.get_item('target_path'), 'rgba')
-		target_vision_mask = extract_vision_mask(target_vision_frame)
+		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, [ target_vision_frame ], preview_mode, preview_resolution)
-		target_vision_frame = merge_vision_mask(target_vision_frame, target_vision_mask)
-		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frame, preview_mode, preview_resolution)
 		preview_vision_frame = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGRA2RGBA)
 		return gradio.Image(value = preview_vision_frame, elem_classes = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ])
 
 	if is_video(state_manager.get_item('target_path')):
 		reference_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))
-		temp_vision_frame = read_video_frame(state_manager.get_item('target_path'), frame_number)
+		target_vision_frames = select_video_frames(state_manager.get_item('target_path'), frame_number, state_manager.get_item('target_frame_amount'))
-		temp_vision_mask = extract_vision_mask(temp_vision_frame)
+		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frames, preview_mode, preview_resolution)
-		temp_vision_frame = merge_vision_mask(temp_vision_frame, temp_vision_mask)
-		preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, temp_vision_frame, preview_mode, preview_resolution)
 		preview_vision_frame = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGRA2RGBA)
 		return gradio.Image(value = preview_vision_frame, elem_classes = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ])
 	return gradio.Image(value = None, elem_classes = None)
 
 
 def clear_and_update_preview_image(preview_mode : PreviewMode, preview_resolution : str, frame_number : int = 0) -> gradio.Image:
-	clear_static_faces()
+	clear_faces()
 	return update_preview_image(preview_mode, preview_resolution, frame_number)
 
 
-def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], source_audio_frame : AudioFrame, source_voice_frame : AudioFrame, target_vision_frame : VisionFrame, preview_mode : PreviewMode, preview_resolution : str) -> VisionFrame:
+def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], source_audio_frame : AudioFrame, source_voice_frame : AudioFrame, target_vision_frames : List[VisionFrame], preview_mode : PreviewMode, preview_resolution : str) -> VisionFrame:
+	target_vision_frame = get_middle(target_vision_frames)
 	target_vision_frame = restrict_frame(target_vision_frame, unpack_resolution(preview_resolution))
+	temp_vision_mask = extract_vision_mask(target_vision_frame)
+	target_vision_frame = merge_vision_mask(target_vision_frame, temp_vision_mask)
+	target_vision_frames = [ restrict_frame(vision_frame, unpack_resolution(preview_resolution))[:, :, :3] for vision_frame in target_vision_frames ]
 	temp_vision_frame = target_vision_frame.copy()
-	temp_vision_mask = extract_vision_mask(temp_vision_frame)
 
 	if analyse_frame(target_vision_frame[:, :, :3]):
 		if preview_mode == 'frame-by-frame':
@@ -229,7 +233,7 @@ def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_fr
 			return numpy.hstack((temp_vision_frame, temp_vision_frame))
 
 		if preview_mode == 'face-by-face':
-			target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, target_vision_frame[:, :, :3], temp_vision_frame[:, :, :3])
+			target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, source_vision_frames, target_vision_frame[:, :, :3], temp_vision_frame[:, :, :3])
 			target_crop_vision_frame = obscure_frame(target_crop_vision_frame)
 			output_crop_vision_frame = obscure_frame(output_crop_vision_frame)
 			return numpy.hstack((target_crop_vision_frame, output_crop_vision_frame))
@@ -247,7 +251,7 @@ def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_fr
 				'source_audio_frame': source_audio_frame,
 				'source_voice_frame': source_voice_frame,
 				'source_vision_frames': source_vision_frames,
-				'target_vision_frame': target_vision_frame[:, :, :3],
+				'target_vision_frames': target_vision_frames,
 				'temp_vision_frame': temp_vision_frame[:, :, :3],
 				'temp_vision_mask': temp_vision_mask
 			})
@@ -259,14 +263,14 @@ def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_fr
 		return numpy.hstack((target_vision_frame, temp_vision_frame))
 
 	if preview_mode == 'face-by-face':
-		target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, target_vision_frame, temp_vision_frame)
+		target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, source_vision_frames, target_vision_frame, temp_vision_frame)
 		return numpy.hstack((target_crop_vision_frame, output_crop_vision_frame))
 
 	return temp_vision_frame
 
 
-def create_face_by_face(reference_vision_frame : VisionFrame, target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> Tuple[VisionFrame, VisionFrame]:
+def create_face_by_face(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> Tuple[VisionFrame, VisionFrame]:
-	target_faces = select_faces(reference_vision_frame[:, :, :3], target_vision_frame[:, :, :3])
+	target_faces = select_faces(reference_vision_frame[:, :, :3], source_vision_frames, [ target_vision_frame[:, :, :3] ])
 	target_face = get_one_face(target_faces)
 
 	if target_face:
@@ -296,7 +300,7 @@ def extract_crop_frame(vision_frame : VisionFrame, face : Face) -> Optional[Visi
 
 
 def prepare_output_frame(target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame, temp_vision_mask : Mask) -> VisionFrame:
-	temp_vision_mask = temp_vision_mask.clip(state_manager.get_item('background_remover_color')[-1], 255)
+	temp_vision_mask = temp_vision_mask.clip(state_manager.get_item('background_remover_fill_color')[-1], 255)
 	temp_vision_frame = merge_vision_mask(temp_vision_frame, temp_vision_mask)
 	temp_vision_frame = cv2.resize(temp_vision_frame, target_vision_frame.shape[1::-1])
 	return temp_vision_frame

facefusion/uis/components/target.py

@@ -3,7 +3,7 @@ from typing import Optional, Tuple
 import gradio
 
 from facefusion import state_manager, translator
-from facefusion.face_store import clear_static_faces
+from facefusion.face_store import clear_faces
 from facefusion.filesystem import is_image, is_video
 from facefusion.uis.core import register_ui_component
 from facefusion.uis.types import ComponentOptions, File
@@ -51,7 +51,7 @@ def listen() -> None:
 
 
 def update(file : File) -> Tuple[gradio.Image, gradio.Video]:
-	clear_static_faces()
+	clear_faces()
 
 	if file and is_image(file.name):
 		state_manager.set_item('target_path', file.name)

facefusion/uis/components/trim_frame.py

@@ -3,7 +3,7 @@ from typing import Optional, Tuple
 from gradio_rangeslider import RangeSlider
 
 from facefusion import state_manager, translator
-from facefusion.face_store import clear_static_faces
+from facefusion.face_store import clear_faces
 from facefusion.filesystem import is_video
 from facefusion.uis.core import get_ui_components
 from facefusion.uis.types import ComponentOptions
@@ -53,7 +53,7 @@ def remote_update() -> RangeSlider:
 
 
 def update_trim_frame(trim_frame : Tuple[float, float]) -> None:
-	clear_static_faces()
+	clear_faces()
 	trim_frame_start, trim_frame_end = trim_frame
 	video_frame_total = count_video_frame_total(state_manager.get_item('target_path'))
 	trim_frame_start = int(trim_frame_start) if trim_frame_start > 0 else None

facefusion/uis/components/voice_extractor.py

@@ -1,4 +1,4 @@
-from typing import Optional
+from typing import List, Optional
 
 import gradio
 
@@ -6,7 +6,7 @@ import facefusion.choices
 from facefusion import state_manager, translator, voice_extractor
 from facefusion.filesystem import is_video
 from facefusion.types import VoiceExtractorModel
-from facefusion.uis.core import get_ui_components, register_ui_component
+from facefusion.uis.core import get_ui_component, get_ui_components, register_ui_component
 
 VOICE_EXTRACTOR_MODEL_DROPDOWN : Optional[gradio.Dropdown] = None
 
@@ -14,11 +14,12 @@ VOICE_EXTRACTOR_MODEL_DROPDOWN : Optional[gradio.Dropdown] = None
 def render() -> None:
 	global VOICE_EXTRACTOR_MODEL_DROPDOWN
 
+	has_lip_syncer = 'lip_syncer' in state_manager.get_item('processors')
 	VOICE_EXTRACTOR_MODEL_DROPDOWN = gradio.Dropdown(
 		label = translator.get('uis.voice_extractor_model_dropdown'),
 		choices = facefusion.choices.voice_extractor_models,
 		value = state_manager.get_item('voice_extractor_model'),
-		visible = is_video(state_manager.get_item('target_path'))
+		visible = is_video(state_manager.get_item('target_path')) and has_lip_syncer
 	)
 	register_ui_component('voice_extractor_model_dropdown', VOICE_EXTRACTOR_MODEL_DROPDOWN)
 
@@ -26,17 +27,22 @@ def render() -> None:
 def listen() -> None:
 	VOICE_EXTRACTOR_MODEL_DROPDOWN.change(update_voice_extractor_model, inputs = VOICE_EXTRACTOR_MODEL_DROPDOWN, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)
 
-	for ui_component in get_ui_components(
+	processors_checkbox_group = get_ui_component('processors_checkbox_group')
-	[
+	if processors_checkbox_group:
-		'target_image',
+		processors_checkbox_group.change(remote_update, inputs = processors_checkbox_group, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)
-		'target_video'
+
-	]):
+		for ui_component in get_ui_components(
-		for method in [ 'change', 'clear' ]:
+		[
-			getattr(ui_component, method)(remote_update, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)
+			'target_image',
+			'target_video'
+		]):
+			for method in [ 'change', 'clear' ]:
+				getattr(ui_component, method)(remote_update, inputs = processors_checkbox_group, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)
 
 
-def remote_update() -> gradio.Dropdown:
+def remote_update(processors : List[str]) -> gradio.Dropdown:
-	if is_video(state_manager.get_item('target_path')):
+	has_lip_syncer = 'lip_syncer' in processors
+	if is_video(state_manager.get_item('target_path')) and has_lip_syncer:
 		return gradio.Dropdown(visible = True)
 	return gradio.Dropdown(visible = False)
 

facefusion/uis/components/webcam.py

@@ -10,7 +10,7 @@ from facefusion.streamer import multi_process_capture, open_stream
 from facefusion.types import Fps, VisionFrame, WebcamMode
 from facefusion.uis.core import get_ui_component
 from facefusion.uis.types import File
-from facefusion.vision import unpack_resolution
+from facefusion.vision import fit_cover_frame, unpack_resolution
 
 SOURCE_FILE : Optional[gradio.File] = None
 WEBCAM_IMAGE : Optional[gradio.Image] = None
@@ -90,7 +90,7 @@ def start(webcam_device_id : int, webcam_mode : WebcamMode, webcam_resolution :
 	stream = None
 
 	if webcam_mode in [ 'udp', 'v4l2' ]:
-		stream = open_stream(webcam_mode, webcam_resolution, webcam_fps) # type:ignore[arg-type]
+		stream = open_stream(webcam_mode, webcam_resolution, webcam_fps) #type:ignore[arg-type]
 	webcam_width, webcam_height = unpack_resolution(webcam_resolution)
 
 	if camera_capture and camera_capture.isOpened():
@@ -98,14 +98,15 @@ def start(webcam_device_id : int, webcam_mode : WebcamMode, webcam_resolution :
 		camera_capture.set(cv2.CAP_PROP_FRAME_HEIGHT, webcam_height)
 		camera_capture.set(cv2.CAP_PROP_FPS, webcam_fps)
 
-		for capture_frame in multi_process_capture(camera_capture, webcam_fps):
+		for capture_vision_frame in multi_process_capture(camera_capture, webcam_fps):
-			capture_frame = cv2.cvtColor(capture_frame, cv2.COLOR_BGR2RGB)
+			capture_vision_frame = cv2.cvtColor(capture_vision_frame, cv2.COLOR_BGR2RGB)
+			capture_vision_frame = fit_cover_frame(capture_vision_frame, (webcam_width, webcam_height))
 
 			if webcam_mode == 'inline':
-				yield capture_frame
+				yield capture_vision_frame
-			else:
+			if webcam_mode in [ 'udp', 'v4l2' ]:
 				try:
-					stream.stdin.write(capture_frame.tobytes())
+					stream.stdin.write(capture_vision_frame.tobytes())
 				except Exception:
 					pass
 

facefusion/uis/core.py

@@ -1,4 +1,5 @@
 import importlib
+import logging
 import os
 import warnings
 from types import ModuleType
@@ -72,6 +73,7 @@ def init() -> None:
 	os.environ['GRADIO_ANALYTICS_ENABLED'] = '0'
 	os.environ['GRADIO_TEMP_DIR'] = os.path.join(state_manager.get_item('temp_path'), 'gradio')
 
+	logging.getLogger('asyncio').setLevel(logging.CRITICAL)
 	warnings.filterwarnings('ignore', category = UserWarning, module = 'gradio')
 	gradio.processing_utils._check_allowed = uis_overrides.mock
 	gradio.processing_utils.convert_video_to_playable_mp4 = uis_overrides.convert_video_to_playable_mp4