github/DeepFaceLab
1
0
Fork 0
mirror of https://github.com/iperov/DeepFaceLab.git synced 2025-07-07 05:22:06 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

SAEHD: added option Enable random warp of samples, default is on

Browse source 
Random warp is required to generalize facial expressions of both faces. When the face is trained enough, you can disable it to get extra sharpness for less amount of iterations.
This commit is contained in:
Colombo 2019-10-12 10:31:50 +04:00
parent e15f846d08
commit 92f14dee70
4 changed files with 33 additions and 29 deletions
Download patch file Download diff file Expand all files Collapse all files

16
models/Model_SAEHD/Model.py
View file

@ -63,6 +63,9 @@ class SAEv2Model(ModelBase):
default_face_style_power = 0.0
default_bg_style_power = 0.0
if is_first_run or ask_override:
default_random_warp = self.options.get('random_warp', True)
self.options['random_warp'] = io.input_str (f"Enable random warp of samples? ( y/n, ?:help skip:{yn_str[default_random_warp]}) : ", default_random_warp, help_message="Random warp is required to generalize facial expressions of both faces. When the face is trained enough, you can disable it to get extra sharpness for less amount of iterations.")
default_face_style_power = default_face_style_power if is_first_run else self.options.get('face_style_power', default_face_style_power)
self.options['face_style_power'] = np.clip ( io.input_number("Face style power ( 0.0 .. 100.0 ?:help skip:%.2f) : " % (default_face_style_power), default_face_style_power,
help_message="Learn to transfer face style details such as light and color conditions. Warning: Enable it only after 10k iters, when predicted face is clear enough to start learn style. Start from 0.1 value and check history changes. Enabling this option increases the chance of model collapse."), 0.0, 100.0 )
@ -84,6 +87,7 @@ class SAEv2Model(ModelBase):
self.options['clipgrad'] = False
else:
self.options['random_warp'] = self.options.get('random_warp', True)
self.options['face_style_power'] = self.options.get('face_style_power', default_face_style_power)
self.options['bg_style_power'] = self.options.get('bg_style_power', default_bg_style_power)
self.options['apply_random_ct'] = self.options.get('apply_random_ct', False)
@ -452,7 +456,7 @@ class SAEv2Model(ModelBase):
self.src_dst_opt = RMSprop(lr=5e-5, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1)
self.src_dst_mask_opt = RMSprop(lr=5e-5, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1)
self.D_opt = RMSprop(lr=5e-5, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1)
src_loss = K.mean ( 10*dssim(kernel_size=int(resolution/11.6),max_value=1.0)( target_src_masked_opt, pred_src_src_masked_opt) )
src_loss += K.mean ( 10*K.square( target_src_masked_opt - pred_src_src_masked_opt ) )
@ -517,7 +521,7 @@ class SAEv2Model(ModelBase):
face_type = t.FACE_TYPE_MID_FULL
elif self.options['face_type'] == 'f':
face_type = t.FACE_TYPE_FULL
t_mode_bgr = t.MODE_BGR if not self.pretrain else t.MODE_BGR_SHUFFLE
training_data_src_path = self.training_data_src_path
@ -528,20 +532,22 @@ class SAEv2Model(ModelBase):
training_data_src_path = self.pretraining_data_path
training_data_dst_path = self.pretraining_data_path
sort_by_yaw = False
t_img_warped = t.IMG_WARPED_TRANSFORMED if self.options['random_warp'] else t.IMG_TRANSFORMED
self.set_training_data_generators ([
SampleGeneratorFace(training_data_src_path, sort_by_yaw_target_samples_path=training_data_dst_path if sort_by_yaw else None,
random_ct_samples_path=training_data_dst_path if apply_random_ct else None,
debug=self.is_debug(), batch_size=self.batch_size,
sample_process_options=SampleProcessor.Options(random_flip=self.random_flip, scale_range=np.array([-0.05, 0.05])+self.src_scale_mod / 100.0 ),
output_sample_types = [ {'types' : (t.IMG_WARPED_TRANSFORMED, face_type, t_mode_bgr), 'resolution':resolution, 'apply_ct': apply_random_ct},
output_sample_types = [ {'types' : (t_img_warped, face_type, t_mode_bgr), 'resolution':resolution, 'apply_ct': apply_random_ct},
{'types' : (t.IMG_TRANSFORMED, face_type, t_mode_bgr), 'resolution': resolution, 'apply_ct': apply_random_ct },
{'types' : (t.IMG_TRANSFORMED, face_type, t.MODE_M), 'resolution': resolution } ]
),
),
SampleGeneratorFace(training_data_dst_path, debug=self.is_debug(), batch_size=self.batch_size,
sample_process_options=SampleProcessor.Options(random_flip=self.random_flip, ),
output_sample_types = [ {'types' : (t.IMG_WARPED_TRANSFORMED, face_type, t_mode_bgr), 'resolution':resolution},
output_sample_types = [ {'types' : (t_img_warped, face_type, t_mode_bgr), 'resolution':resolution},
{'types' : (t.IMG_TRANSFORMED, face_type, t_mode_bgr), 'resolution': resolution},
{'types' : (t.IMG_TRANSFORMED, face_type, t.MODE_M), 'resolution': resolution} ])
])