-
Notifications
You must be signed in to change notification settings - Fork 40
Expand file tree
/
Copy pathdebug_log.txt
More file actions
364 lines (364 loc) · 33.5 KB
/
debug_log.txt
File metadata and controls
364 lines (364 loc) · 33.5 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
2025-09-19 10:31:41,930 - __main__ - INFO -
================================================================================
2025-09-19 10:31:41,931 - __main__ - INFO - STARTING DEBUG FACE SWAP PIPELINE
2025-09-19 10:31:41,931 - __main__ - INFO - ================================================================================
2025-09-19 10:31:41,931 - __main__ - INFO -
Loading volumetric avatar model...
2025-09-19 10:31:49,559 - __main__ - INFO - Model weights loaded successfully
2025-09-19 10:31:50,450 - __main__ - INFO -
Loading images...
2025-09-19 10:31:50,568 - __main__ - INFO - Source image loaded: torch.Size([1, 3, 512, 512])
2025-09-19 10:31:50,568 - __main__ - INFO - Target image loaded: torch.Size([1, 3, 512, 512])
2025-09-19 10:31:50,568 - __main__ - INFO -
################################################################################
2025-09-19 10:31:50,568 - __main__ - INFO - PHASE 1: PROCESSING SOURCE IMAGE (IDENTITY EXTRACTION)
2025-09-19 10:31:50,568 - __main__ - INFO - ################################################################################
2025-09-19 10:31:50,568 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,568 - __main__ - INFO - STEP 1: Face IDT Forward
2025-09-19 10:31:50,568 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,592 - __main__ - DEBUG - input: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=-0.4038, std=0.4596
2025-09-19 10:31:50,737 - __main__ - DEBUG - face_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.int64
2025-09-19 10:31:50,737 - __main__ - DEBUG - cloth_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.int64
2025-09-19 10:31:50,742 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,742 - __main__ - INFO - STEP 2: Masking Source Image
2025-09-19 10:31:50,742 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,742 - __main__ - DEBUG - masked_source: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=-0.2083, std=0.3940
2025-09-19 10:31:50,742 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,742 - __main__ - INFO - STEP 3: Identity Embedder
2025-09-19 10:31:50,742 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,743 - __main__ - DEBUG - masked_input: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=-0.2083, std=0.3940
2025-09-19 10:31:50,786 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32, range=[-2.5306, 2.1158], mean=-0.0056, std=0.4666
2025-09-19 10:31:50,786 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,786 - __main__ - INFO - STEP 4: Head Pose Regressor
2025-09-19 10:31:50,786 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,787 - __main__ - DEBUG - input: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=-0.4038, std=0.4596
2025-09-19 10:31:50,802 - __main__ - DEBUG - theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32, range=[-0.1151, 1.0000], mean=0.1820, std=0.3541
2025-09-19 10:31:50,838 - __main__ - DEBUG - scale=tensor([[0.6465, 0.7250, 0.6786]], device='cuda:0'), rotation=tensor([[-0.0089, -0.1232, 0.1605]], device='cuda:0'), translation=tensor([[ 0.0101, -0.1019, -0.0373]], device='cuda:0')
2025-09-19 10:31:50,838 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,838 - __main__ - INFO - STEP 5: Creating Source Data Dict
2025-09-19 10:31:50,838 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,838 - __main__ - DEBUG -
=== source_data_dict ===
2025-09-19 10:31:50,838 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,838 - __main__ - INFO - STEP 6: Expression Embedder
2025-09-19 10:31:50,838 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,838 - __main__ - DEBUG -
=== input_data_dict ===
2025-09-19 10:31:50,838 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,838 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG -
=== output_data_dict ===
2025-09-19 10:31:50,868 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - pred_source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - pred_target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - source_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - target_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - align_warp: shape=[2, 128, 128, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - align_warp_full: shape=[2, 512, 512, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - source_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,868 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - source_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32, range=[-1.6122, 2.0202], mean=0.0039, std=0.5080
2025-09-19 10:31:50,869 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32, range=[-1.6122, 2.0202], mean=0.0039, std=0.5080
2025-09-19 10:31:50,869 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,869 - __main__ - INFO - STEP 7: Predict Embed
2025-09-19 10:31:50,869 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,869 - __main__ - DEBUG -
=== input_data_dict ===
2025-09-19 10:31:50,869 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - pred_source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - pred_target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - source_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - target_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - align_warp: shape=[2, 128, 128, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - align_warp_full: shape=[2, 512, 512, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - source_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,869 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,871 - __main__ - DEBUG - Source warp embed dict:
2025-09-19 10:31:50,871 - __main__ - DEBUG - source_warp[orig]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,871 - __main__ - DEBUG - source_warp[orig_d]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,871 - __main__ - DEBUG - source_warp[ada_v]: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,871 - __main__ - DEBUG - Target warp embed dict:
2025-09-19 10:31:50,871 - __main__ - DEBUG - target_warp[orig]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,871 - __main__ - DEBUG - target_warp[orig_d]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,871 - __main__ - DEBUG - target_warp[ada_v]: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,871 - __main__ - DEBUG -
=== embed_dict ===
2025-09-19 10:31:50,871 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,871 - __main__ - INFO - STEP 8: XY Warp Generator
2025-09-19 10:31:50,871 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,884 - __main__ - DEBUG - xy_warp: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-1.3948, 2.0000], mean=0.0049, std=0.7132
2025-09-19 10:31:50,885 - __main__ - DEBUG - xy_conf: shape=[1, 3, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=0.0049, std=0.2581
2025-09-19 10:31:50,885 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,885 - __main__ - INFO - STEP 9: Local Encoder
2025-09-19 10:31:50,885 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,885 - __main__ - DEBUG - masked_input: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=-0.2083, std=0.3940
2025-09-19 10:31:50,906 - __main__ - DEBUG - source_latents: shape=[1, 1536, 64, 64], device=cuda:0, dtype=torch.float32, range=[-43.6333, 36.9271], mean=0.1253, std=5.8706
2025-09-19 10:31:50,906 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,906 - __main__ - INFO - STEP 10: Creating Source Volume
2025-09-19 10:31:50,906 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,906 - __main__ - DEBUG - Volume dimensions: c=96, d=16, s=64
2025-09-19 10:31:50,906 - __main__ - DEBUG - source_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-43.6333, 36.9271], mean=0.1253, std=5.8706
2025-09-19 10:31:50,906 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,906 - __main__ - INFO - STEP 11: Volume Source Network
2025-09-19 10:31:50,906 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,906 - __main__ - DEBUG - input_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-43.6333, 36.9271], mean=0.1253, std=5.8706
2025-09-19 10:31:50,911 - __main__ - DEBUG - processed_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-49.9362, 44.7376], mean=0.1965, std=6.2137
2025-09-19 10:31:50,911 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,911 - __main__ - INFO - STEP 12: Creating Source Rotation Warp
2025-09-19 10:31:50,911 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,911 - __main__ - DEBUG - identity_grid: shape=[1, 65536, 4], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=0.2500, std=0.6737
2025-09-19 10:31:50,912 - __main__ - DEBUG - source_rotation_warp: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-0.9060, 0.8209], mean=-0.0312, std=0.4089
2025-09-19 10:31:50,912 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,912 - __main__ - INFO - STEP 13: Grid Sample (source rotation)
2025-09-19 10:31:50,912 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,912 - __main__ - DEBUG - input_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-49.9362, 44.7376], mean=0.1965, std=6.2137
2025-09-19 10:31:50,912 - __main__ - DEBUG - warp_field: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-0.9060, 0.8209], mean=-0.0312, std=0.4089
2025-09-19 10:31:50,913 - __main__ - DEBUG - warped_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-43.2830, 37.7650], mean=0.2225, std=5.8920
2025-09-19 10:31:50,913 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,913 - __main__ - INFO - STEP 14: Grid Sample (source XY)
2025-09-19 10:31:50,913 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,913 - __main__ - DEBUG - input_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-43.2830, 37.7650], mean=0.2225, std=5.8920
2025-09-19 10:31:50,913 - __main__ - DEBUG - warp_field: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-1.3948, 2.0000], mean=0.0049, std=0.7132
2025-09-19 10:31:50,913 - __main__ - DEBUG - warped_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-41.4566, 36.6019], mean=0.2094, std=4.4236
2025-09-19 10:31:50,913 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,913 - __main__ - INFO - STEP 15: Volume Process Network
2025-09-19 10:31:50,913 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,914 - __main__ - DEBUG - input_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-41.4566, 36.6019], mean=0.2094, std=4.4236
2025-09-19 10:31:50,914 - __main__ - DEBUG -
=== embed_dict ===
2025-09-19 10:31:50,954 - __main__ - DEBUG - processed_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-2.9867, 2.4477], mean=0.0034, std=0.1904
2025-09-19 10:31:50,954 - __main__ - INFO -
>>> SOURCE PROCESSING COMPLETE <<<
2025-09-19 10:31:50,954 - __main__ - INFO - Identity embedding shape: torch.Size([1, 512, 4, 4])
2025-09-19 10:31:50,954 - __main__ - INFO - Processed canonical shape: torch.Size([1, 96, 16, 64, 64])
2025-09-19 10:31:50,954 - __main__ - INFO -
################################################################################
2025-09-19 10:31:50,954 - __main__ - INFO - PHASE 2: PROCESSING TARGET IMAGE (EXPRESSION APPLICATION)
2025-09-19 10:31:50,954 - __main__ - INFO - ################################################################################
2025-09-19 10:31:50,954 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,954 - __main__ - INFO - STEP 16: Face IDT Forward
2025-09-19 10:31:50,954 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,954 - __main__ - DEBUG - input: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-0.9922, 1.0000], mean=-0.1945, std=0.4144
2025-09-19 10:31:50,957 - __main__ - DEBUG - face_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.int64
2025-09-19 10:31:50,957 - __main__ - DEBUG - cloth_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.int64
2025-09-19 10:31:50,957 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,957 - __main__ - INFO - STEP 17: Head Pose Regressor
2025-09-19 10:31:50,957 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,958 - __main__ - DEBUG - input: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-0.9922, 1.0000], mean=-0.1945, std=0.4144
2025-09-19 10:31:50,959 - __main__ - DEBUG - theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32, range=[-0.0629, 1.0000], mean=0.1981, std=0.3764
2025-09-19 10:31:50,961 - __main__ - DEBUG - scale=tensor([[0.7080, 0.8128, 0.7647]], device='cuda:0'), rotation=tensor([[-0.0245, 0.0359, -0.0242]], device='cuda:0'), translation=tensor([[ 0.0058, -0.0759, -0.0638]], device='cuda:0')
2025-09-19 10:31:50,961 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,961 - __main__ - INFO - STEP 18: Creating Target Data Dict (with Source Identity)
2025-09-19 10:31:50,961 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,961 - __main__ - DEBUG -
=== target_data_dict ===
2025-09-19 10:31:50,961 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,961 - __main__ - INFO - STEP 19: Expression Embedder
2025-09-19 10:31:50,961 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,961 - __main__ - DEBUG -
=== input_data_dict ===
2025-09-19 10:31:50,961 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,961 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG -
=== output_data_dict ===
2025-09-19 10:31:50,964 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - pred_source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - pred_target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - source_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - target_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - align_warp: shape=[2, 128, 128, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - align_warp_full: shape=[2, 512, 512, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - source_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,964 - __main__ - DEBUG - source_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32, range=[-2.1612, 1.7617], mean=0.0099, std=0.5643
2025-09-19 10:31:50,965 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32, range=[-2.1612, 1.7617], mean=0.0099, std=0.5643
2025-09-19 10:31:50,965 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,965 - __main__ - INFO - STEP 20: Predict Embed
2025-09-19 10:31:50,965 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,965 - __main__ - DEBUG -
=== input_data_dict ===
2025-09-19 10:31:50,965 - __main__ - DEBUG - source_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - source_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - target_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - target_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - idt_embed: shape=[1, 512, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - pred_source_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - pred_target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - source_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - target_img_align: shape=[1, 3, 128, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - align_warp: shape=[2, 128, 128, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - align_warp_full: shape=[2, 512, 512, 2], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - source_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,965 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,966 - __main__ - DEBUG - Source warp embed dict:
2025-09-19 10:31:50,966 - __main__ - DEBUG - source_warp[orig]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,966 - __main__ - DEBUG - source_warp[orig_d]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,966 - __main__ - DEBUG - source_warp[ada_v]: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,966 - __main__ - DEBUG - Target warp embed dict:
2025-09-19 10:31:50,966 - __main__ - DEBUG - target_warp[orig]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,966 - __main__ - DEBUG - target_warp[orig_d]: shape=[1, 512, 16], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,966 - __main__ - DEBUG - target_warp[ada_v]: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,966 - __main__ - DEBUG -
=== embed_dict ===
2025-09-19 10:31:50,966 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,966 - __main__ - INFO - STEP 21: UV Warp Generator
2025-09-19 10:31:50,966 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,970 - __main__ - DEBUG - uv_warp: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-1.4054, 1.9993], mean=0.1466, std=0.7513
2025-09-19 10:31:50,970 - __main__ - DEBUG - uv_conf: shape=[1, 3, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-1.0000, 1.0000], mean=0.1466, std=0.3403
2025-09-19 10:31:50,970 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,970 - __main__ - INFO - STEP 22: Grid Sample (target UV)
2025-09-19 10:31:50,970 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,971 - __main__ - DEBUG - input_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-2.9867, 2.4477], mean=0.0034, std=0.1904
2025-09-19 10:31:50,971 - __main__ - DEBUG - warp_field: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-1.4054, 1.9993], mean=0.1466, std=0.7513
2025-09-19 10:31:50,971 - __main__ - DEBUG - warped_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-2.6772, 1.8829], mean=0.0046, std=0.1081
2025-09-19 10:31:50,971 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,971 - __main__ - INFO - STEP 23: Creating Target Rotation Warp
2025-09-19 10:31:50,971 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,971 - __main__ - DEBUG - target_rotation_warp: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-0.9142, 0.7883], mean=-0.0364, std=0.4556
2025-09-19 10:31:50,971 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,971 - __main__ - INFO - STEP 24: Grid Sample (target rotation)
2025-09-19 10:31:50,972 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,972 - __main__ - DEBUG - input_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-2.6772, 1.8829], mean=0.0046, std=0.1081
2025-09-19 10:31:50,972 - __main__ - DEBUG - warp_field: shape=[1, 16, 64, 64, 3], device=cuda:0, dtype=torch.float32, range=[-0.9142, 0.7883], mean=-0.0364, std=0.4556
2025-09-19 10:31:50,972 - __main__ - DEBUG - warped_volume: shape=[1, 96, 16, 64, 64], device=cuda:0, dtype=torch.float32, range=[-2.0513, 1.3437], mean=0.0071, std=0.1258
2025-09-19 10:31:50,972 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,972 - __main__ - INFO - STEP 25: Preparing Decoder Input
2025-09-19 10:31:50,972 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,973 - __main__ - DEBUG - target_latent_feats: shape=[1, 1536, 64, 64], device=cuda:0, dtype=torch.float32, range=[-2.0513, 1.3437], mean=0.0071, std=0.1258
2025-09-19 10:31:50,973 - __main__ - DEBUG -
=== decode_dict ===
2025-09-19 10:31:50,973 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,973 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,973 - __main__ - INFO -
============================================================
2025-09-19 10:31:50,973 - __main__ - INFO - STEP 26: Decoder Network
2025-09-19 10:31:50,973 - __main__ - INFO - ============================================================
2025-09-19 10:31:50,973 - __main__ - DEBUG -
=== data_dict ===
2025-09-19 10:31:50,973 - __main__ - DEBUG - target_theta: shape=[1, 4, 4], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,973 - __main__ - DEBUG - target_pose_embed: shape=[1, 128], device=cuda:0, dtype=torch.float32
2025-09-19 10:31:50,973 - __main__ - DEBUG -
=== embed_dict ===
2025-09-19 10:31:50,973 - __main__ - DEBUG - latent_feats: shape=[1, 1536, 64, 64], device=cuda:0, dtype=torch.float32, range=[-2.0513, 1.3437], mean=0.0071, std=0.1258
2025-09-19 10:31:50,973 - __main__ - DEBUG - stage_two=True
2025-09-19 10:31:51,001 - __main__ - DEBUG - generated_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[0.0000, 1.0000], mean=0.2411, std=0.3128
2025-09-19 10:31:51,001 - __main__ - DEBUG - deep_features: shape=[1, 512, 64, 64], device=cuda:0, dtype=torch.float32, range=[-38.4213, 25.2028], mean=-0.0059, std=0.5857
2025-09-19 10:31:51,002 - __main__ - DEBUG - img_features: shape=[1, 128, 512, 512], device=cuda:0, dtype=torch.float32, range=[-206.2700, 94.2017], mean=-0.0297, std=2.0088
2025-09-19 10:31:51,002 - __main__ - INFO -
============================================================
2025-09-19 10:31:51,002 - __main__ - INFO - STEP 27: Face IDT Forward
2025-09-19 10:31:51,002 - __main__ - INFO - ============================================================
2025-09-19 10:31:51,002 - __main__ - DEBUG - input: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[0.0000, 1.0000], mean=0.2411, std=0.3128
2025-09-19 10:31:51,005 - __main__ - DEBUG - face_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.int64
2025-09-19 10:31:51,005 - __main__ - DEBUG - cloth_mask: shape=[1, 1, 512, 512], device=cuda:0, dtype=torch.int64
2025-09-19 10:31:51,005 - __main__ - INFO -
============================================================
2025-09-19 10:31:51,005 - __main__ - INFO - STEP 28: Final Compositing
2025-09-19 10:31:51,005 - __main__ - INFO - ============================================================
2025-09-19 10:31:51,006 - __main__ - DEBUG - final_img: shape=[1, 3, 512, 512], device=cuda:0, dtype=torch.float32, range=[-0.9529, 1.0000], mean=0.0041, std=0.3580
2025-09-19 10:31:51,006 - __main__ - INFO -
################################################################################
2025-09-19 10:31:51,006 - __main__ - INFO - PHASE 3: SAVING RESULTS
2025-09-19 10:31:51,006 - __main__ - INFO - ################################################################################
2025-09-19 10:31:51,669 - __main__ - INFO - Saved result to debug_face_swap_result.png
2025-09-19 10:31:51,732 - __main__ - INFO - Saved final output to debug_final_output.png
2025-09-19 10:31:51,732 - __main__ - INFO -
================================================================================
2025-09-19 10:31:51,732 - __main__ - INFO - DEBUG PIPELINE COMPLETE
2025-09-19 10:31:51,732 - __main__ - INFO - ================================================================================
2025-09-19 10:31:51,732 - __main__ - INFO -
Summary:
2025-09-19 10:31:51,732 - __main__ - INFO - Total steps logged: 28
2025-09-19 10:31:51,732 - __main__ - INFO - Source identity shape: torch.Size([1, 512, 4, 4])
2025-09-19 10:31:51,732 - __main__ - INFO - Final output shape: torch.Size([1, 3, 512, 512])
2025-09-19 10:31:51,732 - __main__ - INFO - Output range: [-0.953, 1.000]
Hybrid stages [True, True, True]
bbbbbbbbbbb
Hybrid stages [True, True, True]
SN applied to local_encoder_nw
SN applied to idt_embedder_nw
SN applied to expression_embedder_nw
SN applied to xy_generator_nw
SN applied to uv_generator_nw
SN applied to warp_embed_head_orig_nw
SN applied to volume_process_nw
SN applied to volume_source_nw
SN applied to decoder_nw
WS applied to local_encoder_nw
WS applied to idt_embedder_nw
WS applied to expression_embedder_nw
WS applied to xy_generator_nw
WS applied to uv_generator_nw
WS applied to warp_embed_head_orig_nw
WS applied to volume_process_nw
WS applied to volume_source_nw
WS applied to decoder_nw