github/DeepFaceLab
1
0
Fork 0
mirror of https://github.com/iperov/DeepFaceLab.git synced 2025-07-06 21:12:07 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

提供自動合併機制。

Browse source
This commit is contained in:
yaoming 2022-10-09 00:48:04 +08:00
parent 335f887ba9
commit a767b6790b
4 changed files with 98 additions and 37 deletions
Download patch file Download diff file Expand all files Collapse all files

23
.vscode/launch.json vendored
View file

@ -38,7 +38,8 @@
"--training-data-src-dir", "D:\\DeepFaceLab\\workspace\\data_src\\aligned", "--training-data-src-dir", "D:\\DeepFaceLab\\workspace\\data_src\\aligned",
"--training-data-dst-dir", "D:\\DeepFaceLab\\workspace\\data_dst\\aligned", "--training-data-dst-dir", "D:\\DeepFaceLab\\workspace\\data_dst\\aligned",
"--model-dir", "D:\\DeepFaceLab\\workspace\\model", "--model-dir", "D:\\DeepFaceLab\\workspace\\model",
"--model", "SAEHD" "--model", "SAEHD",
"--silent-start"
] ]
}, },
{ {
@ -63,6 +64,26 @@
"--force-gpu-idxs", "0" "--force-gpu-idxs", "0"
] ]
}, },
{
"name": "DFL mp4",
"subProcess": true,
"justMyCode": true,
"type": "python",
"request": "launch",
"program": "D:\\DeepFaceLab\\_internal\\DeepFaceLab\\main.py",
"python": "D:\\DeepFaceLab\\_internal\\python-3.6.8\\python.exe",
"cwd": "D:\\DeepFaceLab\\workspace",
"console": "integratedTerminal",
"gevent": true,
"args": ["videoed",
"video-from-sequence",
"--input-dir", "data_dst\\merged",
"--output-file", "result.mp4",
"--reference-file", "data_dst.*",
"--bitrate", "16",
"--include-audio"
]
},
{ {
"name": "Auto DFL", "name": "Auto DFL",
"subProcess": true, "subProcess": true,

5
autodfl.py
View file

@ -2,8 +2,9 @@ import ymauto.MergeDefault as MD
md = MD.MergeArgs("config.json") md = MD.MergeArgs("config.json")
print(md.g("sandy", "name")) print(md.gOrder("mask_mode", "mask_mode_opts"))
print(md.g("beauty", "deep", "gender")) # print(md.g("sandy", "name"))
# print(md.g("beauty", "deep", "gender"))
# print(md.g("version1")) # print(md.g("version1"))
# print(md.g("deep", "method")) # print(md.g("deep", "method"))
# print(md.g("deep", "methoda", "PP")) # print(md.g("deep", "methoda", "PP"))

8
mainscripts/Merger.py
View file

@ -71,19 +71,19 @@ def main (model_class_name=None,
place_model_on_cpu=True, place_model_on_cpu=True,
run_on_cpu=run_on_cpu) run_on_cpu=run_on_cpu)
if md.g(None, "interactive") is None: if md.g("interactive") is None:
is_interactive = io.input_bool ("Use interactive merger?", True) if not io.is_colab() else False is_interactive = io.input_bool ("Use interactive merger?", True) if not io.is_colab() else False
else: else:
is_interactive = md.g(None, "interactive") is_interactive = md.g("interactive")
if not is_interactive: if not is_interactive:
cfg.ask_settings() cfg.ask_settings()
if md.g(None, "subprocess_count") is None: if md.gd("NoSet", "subprocess_count") == "NoSet":
subprocess_count = io.input_int("Number of workers?", max(8, multiprocessing.cpu_count()), subprocess_count = io.input_int("Number of workers?", max(8, multiprocessing.cpu_count()),
valid_range=[1, multiprocessing.cpu_count()], help_message="Specify the number of threads to process. A low value may affect performance. A high value may result in memory error. The value may not be greater than CPU cores." ) valid_range=[1, multiprocessing.cpu_count()], help_message="Specify the number of threads to process. A low value may affect performance. A high value may result in memory error. The value may not be greater than CPU cores." )
else: else:
subprocess_count = md.g(4, "subprocess_count") subprocess_count = md.gd(4, "subprocess_count")
input_path_image_paths = pathex.get_image_paths(input_path) input_path_image_paths = pathex.get_image_paths(input_path)

91
merger/MergerConfig.py
View file

@ -34,11 +34,14 @@ class MergerConfig(object):
#overridable #overridable
def ask_settings(self): def ask_settings(self):
s = """Choose sharpen mode: \n""" if md.g("sharpen_mode") is None:
for key in self.sharpen_dict.keys(): s = """Choose sharpen mode: \n"""
s += f"""({key}) {self.sharpen_dict[key]}\n""" for key in self.sharpen_dict.keys():
io.log_info(s) s += f"""({key}) {self.sharpen_dict[key]}\n"""
self.sharpen_mode = io.input_int ("", 0, valid_list=self.sharpen_dict.keys(), help_message="Enhance details by applying sharpen filter.") io.log_info(s)
self.sharpen_mode = io.input_int ("", 0, valid_list=self.sharpen_dict.keys(), help_message="Enhance details by applying sharpen filter.")
else:
self.sharpen_mode = md.gOrder("sharpen_mode", "sharpen_mode_opts")
if self.sharpen_mode != 0: if self.sharpen_mode != 0:
self.blursharpen_amount = np.clip ( io.input_int ("Choose blur/sharpen amount", 0, add_info="-100..100"), -100, 100 ) self.blursharpen_amount = np.clip ( io.input_int ("Choose blur/sharpen amount", 0, add_info="-100..100"), -100, 100 )
@ -191,15 +194,17 @@ class MergerConfigMasked(MergerConfig):
self.bicubic_degrade_power = np.clip ( self.bicubic_degrade_power+diff, 0, 100) self.bicubic_degrade_power = np.clip ( self.bicubic_degrade_power+diff, 0, 100)
def ask_settings(self): def ask_settings(self):
print("mode: " + md.g('mode'))
s = """Choose mode: \n""" if md.g("mode") is None:
for key in mode_dict.keys(): s = """Choose mode: \n"""
s += f"""({key}) {mode_dict[key]}\n""" for key in mode_dict.keys():
io.log_info(s) s += f"""({key}) {mode_dict[key]}\n"""
mode = io.input_int ("", mode_str_dict.get(self.default_mode, 1) ) io.log_info(s)
mode = io.input_int ("", mode_str_dict.get(self.default_mode, 1) )
self.mode = mode_dict.get (mode, self.default_mode ) self.mode = mode_dict.get (mode, self.default_mode )
else:
self.mode = md.g("mode")
if 'raw' not in self.mode: if 'raw' not in self.mode:
if self.mode == 'hist-match': if self.mode == 'hist-match':
@ -208,31 +213,65 @@ class MergerConfigMasked(MergerConfig):
if self.mode == 'hist-match' or self.mode == 'seamless-hist-match': if self.mode == 'hist-match' or self.mode == 'seamless-hist-match':
self.hist_match_threshold = np.clip ( io.input_int("Hist match threshold", 255, add_info="0..255"), 0, 255) self.hist_match_threshold = np.clip ( io.input_int("Hist match threshold", 255, add_info="0..255"), 0, 255)
s = """Choose mask mode: \n""" if md.g("mask_mode") is None:
for key in mask_mode_dict.keys(): s = """Choose mask mode: \n"""
s += f"""({key}) {mask_mode_dict[key]}\n""" for key in mask_mode_dict.keys():
io.log_info(s) s += f"""({key}) {mask_mode_dict[key]}\n"""
self.mask_mode = io.input_int ("", 1, valid_list=mask_mode_dict.keys() ) io.log_info(s)
self.mask_mode = io.input_int ("", 1, valid_list=mask_mode_dict.keys() )
else:
self.mask_mode = md.gOrder("mask_mode", "mask_mode_opts")
if 'raw' not in self.mode: if 'raw' not in self.mode:
self.erode_mask_modifier = np.clip ( io.input_int ("Choose erode mask modifier", 0, add_info="-400..400"), -400, 400) if md.g("erode_mask_modifier") is None:
self.blur_mask_modifier = np.clip ( io.input_int ("Choose blur mask modifier", 0, add_info="0..400"), 0, 400) self.erode_mask_modifier = np.clip ( io.input_int ("Choose erode mask modifier", 0, add_info="-400..400"), -400, 400)
self.motion_blur_power = np.clip ( io.input_int ("Choose motion blur power", 0, add_info="0..100"), 0, 100) else:
self.erode_mask_modifier = md.g("erode_mask_modifier")
self.output_face_scale = np.clip (io.input_int ("Choose output face scale modifier", 0, add_info="-50..50" ), -50, 50) if md.g("blur_mask_modifier") is None:
self.blur_mask_modifier = np.clip ( io.input_int ("Choose blur mask modifier", 0, add_info="0..400"), 0, 400)
else:
self.erode_mask_modifier = md.g("blur_mask_modifier")
if md.g("motion_blur_power") is None:
self.motion_blur_power = np.clip ( io.input_int ("Choose motion blur power", 0, add_info="0..100"), 0, 100)
else:
self.erode_mask_modifier = md.g("motion_blur_power")
if md.g("output_face_scale") is None:
self.output_face_scale = np.clip (io.input_int ("Choose output face scale modifier", 0, add_info="-50..50" ), -50, 50)
else:
self.output_face_scale = md.g("output_face_scale")
if 'raw' not in self.mode: if 'raw' not in self.mode:
self.color_transfer_mode = io.input_str ( "Color transfer to predicted face", None, valid_list=list(ctm_str_dict.keys())[1:] ) if md.g("color_transfer_mode") is None:
self.color_transfer_mode = io.input_str ( "Color transfer to predicted face", None, valid_list=list(ctm_str_dict.keys())[1:] )
else:
self.color_transfer_mode = md.g("color_transfer_mode")
self.color_transfer_mode = ctm_str_dict[self.color_transfer_mode] self.color_transfer_mode = ctm_str_dict[self.color_transfer_mode]
super().ask_settings() super().ask_settings()
self.super_resolution_power = np.clip ( io.input_int ("Choose super resolution power", 0, add_info="0..100", help_message="Enhance details by applying superresolution network."), 0, 100) if md.g("super_resolution_power") is None:
self.super_resolution_power = np.clip ( io.input_int ("Choose super resolution power", 0, add_info="0..100", help_message="Enhance details by applying superresolution network."), 0, 100)
else:
self.super_resolution_power = md.g("super_resolution_power")
if 'raw' not in self.mode: if 'raw' not in self.mode:
self.image_denoise_power = np.clip ( io.input_int ("Choose image degrade by denoise power", 0, add_info="0..500"), 0, 500) if md.g("image_denoise_power") is None:
self.bicubic_degrade_power = np.clip ( io.input_int ("Choose image degrade by bicubic rescale power", 0, add_info="0..100"), 0, 100) self.image_denoise_power = np.clip ( io.input_int ("Choose image degrade by denoise power", 0, add_info="0..500"), 0, 500)
self.color_degrade_power = np.clip ( io.input_int ("Degrade color power of final image", 0, add_info="0..100"), 0, 100) else:
self.image_denoise_power = md.g("image_denoise_power")
if md.g("bicubic_degrade_power") is None:
self.bicubic_degrade_power = np.clip ( io.input_int ("Choose image degrade by bicubic rescale power", 0, add_info="0..100"), 0, 100)
else:
self.bicubic_degrade_power = md.g("bicubic_degrade_power")
if md.g("color_degrade_power") is None:
self.color_degrade_power = np.clip ( io.input_int ("Degrade color power of final image", 0, add_info="0..100"), 0, 100)
else:
self.color_degrade_power = md.g("color_degrade_power")
io.log_info ("") io.log_info ("")