File size: 22,966 Bytes
11e6f7b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
# SPDX-FileCopyrightText: Copyright (c) 2021-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: LicenseRef-NvidiaProprietary
#
# NVIDIA CORPORATION, its affiliates and licensors retain all intellectual
# property and proprietary rights in and to this material, related
# documentation and any modifications thereto. Any use, reproduction,
# disclosure or distribution of this material and related documentation
# without an express license agreement from NVIDIA CORPORATION or
# its affiliates is strictly prohibited.
"""Discriminator architectures from the paper
"Efficient Geometry-aware 3D Generative Adversarial Networks"."""

import numpy as np
import torch
from utils.torch_utils import persistence
from utils.torch_utils.ops import upfirdn2d
from .networks_stylegan2 import DiscriminatorBlock, MappingNetwork, DiscriminatorEpilogue
from pdb import set_trace as st


@persistence.persistent_class
class SingleDiscriminator(torch.nn.Module):
    def __init__(
            self,
            c_dim,  # Conditioning label (C) dimensionality.
            img_resolution,  # Input resolution.
            img_channels,  # Number of input color channels.
            architecture='resnet',  # Architecture: 'orig', 'skip', 'resnet'.
            channel_base=32768,  # Overall multiplier for the number of channels.
            channel_max=512,  # Maximum number of channels in any layer.
            num_fp16_res=4,  # Use FP16 for the N highest resolutions.
            conv_clamp=256,  # Clamp the output of convolution layers to +-X, None = disable clamping.
            cmap_dim=None,  # Dimensionality of mapped conditioning label, None = default.
            sr_upsample_factor=1,  # Ignored for SingleDiscriminator
            block_kwargs={},  # Arguments for DiscriminatorBlock.
            mapping_kwargs={},  # Arguments for MappingNetwork.
            epilogue_kwargs={},  # Arguments for DiscriminatorEpilogue.
    ):
        super().__init__()
        self.c_dim = c_dim
        self.img_resolution = img_resolution
        self.img_resolution_log2 = int(np.log2(img_resolution))
        self.img_channels = img_channels
        self.block_resolutions = [
            2**i for i in range(self.img_resolution_log2, 2, -1)
        ]
        channels_dict = {
            res: min(channel_base // res, channel_max)
            for res in self.block_resolutions + [4]
        }
        fp16_resolution = max(2**(self.img_resolution_log2 + 1 - num_fp16_res),
                              8)

        if cmap_dim is None:
            cmap_dim = channels_dict[4]
        if c_dim == 0:
            cmap_dim = 0

        common_kwargs = dict(img_channels=img_channels,
                             architecture=architecture,
                             conv_clamp=conv_clamp)
        cur_layer_idx = 0
        for res in self.block_resolutions:
            in_channels = channels_dict[res] if res < img_resolution else 0
            tmp_channels = channels_dict[res]
            out_channels = channels_dict[res // 2]
            use_fp16 = (res >= fp16_resolution)
            block = DiscriminatorBlock(in_channels,
                                       tmp_channels,
                                       out_channels,
                                       resolution=res,
                                       first_layer_idx=cur_layer_idx,
                                       use_fp16=use_fp16,
                                       **block_kwargs,
                                       **common_kwargs)
            setattr(self, f'b{res}', block)
            cur_layer_idx += block.num_layers
        if c_dim > 0:
            self.mapping = MappingNetwork(z_dim=0,
                                          c_dim=c_dim,
                                          w_dim=cmap_dim,
                                          num_ws=None,
                                          w_avg_beta=None,
                                          **mapping_kwargs)
        self.b4 = DiscriminatorEpilogue(channels_dict[4],
                                        cmap_dim=cmap_dim,
                                        resolution=4,
                                        **epilogue_kwargs,
                                        **common_kwargs)

    def forward(self, img, c, update_emas=False, **block_kwargs):
        img = img['image']

        _ = update_emas  # unused
        x = None
        for res in self.block_resolutions:
            block = getattr(self, f'b{res}')
            x, img = block(x, img, **block_kwargs)

        cmap = None
        if self.c_dim > 0:
            cmap = self.mapping(None, c)
        x = self.b4(x, img, cmap)
        return x

    def extra_repr(self):
        return f'c_dim={self.c_dim:d}, img_resolution={self.img_resolution:d}, img_channels={self.img_channels:d}'


#----------------------------------------------------------------------------


def filtered_resizing(image_orig_tensor, size, f, filter_mode='antialiased'):
    if filter_mode == 'antialiased':
        ada_filtered_64 = torch.nn.functional.interpolate(image_orig_tensor,
                                                          size=(size, size),
                                                          mode='bilinear',
                                                          align_corners=False,
                                                          antialias=True)
    elif filter_mode == 'classic':
        ada_filtered_64 = upfirdn2d.upsample2d(image_orig_tensor, f, up=2)
        ada_filtered_64 = torch.nn.functional.interpolate(ada_filtered_64,
                                                          size=(size * 2 + 2,
                                                                size * 2 + 2),
                                                          mode='bilinear',
                                                          align_corners=False)
        ada_filtered_64 = upfirdn2d.downsample2d(ada_filtered_64,
                                                 f,
                                                 down=2,
                                                 flip_filter=True,
                                                 padding=-1)
    elif filter_mode == 'none':
        ada_filtered_64 = torch.nn.functional.interpolate(image_orig_tensor,
                                                          size=(size, size),
                                                          mode='bilinear',
                                                          align_corners=False)
    elif type(filter_mode) == float:
        assert 0 < filter_mode < 1

        filtered = torch.nn.functional.interpolate(image_orig_tensor,
                                                   size=(size, size),
                                                   mode='bilinear',
                                                   align_corners=False,
                                                   antialias=True)
        aliased = torch.nn.functional.interpolate(image_orig_tensor,
                                                  size=(size, size),
                                                  mode='bilinear',
                                                  align_corners=False,
                                                  antialias=False)
        ada_filtered_64 = (1 -
                           filter_mode) * aliased + (filter_mode) * filtered

    return ada_filtered_64


#----------------------------------------------------------------------------


@persistence.persistent_class
class DualDiscriminator(torch.nn.Module):
    def __init__(
            self,
            c_dim,  # Conditioning label (C) dimensionality.
            img_resolution,  # Input resolution.
            img_channels,  # Number of input color channels.
            architecture='resnet',  # Architecture: 'orig', 'skip', 'resnet'.
            channel_base=32768,  # Overall multiplier for the number of channels.
            channel_max=512,  # Maximum number of channels in any layer.
            num_fp16_res=4,  # Use FP16 for the N highest resolutions.
            conv_clamp=256,  # Clamp the output of convolution layers to +-X, None = disable clamping.
            cmap_dim=None,  # Dimensionality of mapped conditioning label, None = default.
            disc_c_noise=0,  # Corrupt camera parameters with X std dev of noise before disc. pose conditioning.
            block_kwargs={},  # Arguments for DiscriminatorBlock.
            mapping_kwargs={},  # Arguments for MappingNetwork.
            epilogue_kwargs={},  # Arguments for DiscriminatorEpilogue.
    ):
        super().__init__()
        # img_channels *= 2
        if img_channels == 3:
            img_channels *= 2

        self.c_dim = c_dim
        self.img_resolution = img_resolution
        self.img_resolution_log2 = int(np.log2(img_resolution))
        self.img_channels = img_channels
        self.block_resolutions = [
            2**i for i in range(self.img_resolution_log2, 2, -1)
        ]
        channels_dict = {
            res: min(channel_base // res, channel_max)
            for res in self.block_resolutions + [4]
        }
        fp16_resolution = max(2**(self.img_resolution_log2 + 1 - num_fp16_res),
                              8)

        if cmap_dim is None:
            cmap_dim = channels_dict[4]
        if c_dim == 0:
            cmap_dim = 0

        common_kwargs = dict(img_channels=img_channels,
                             architecture=architecture,
                             conv_clamp=conv_clamp)
        cur_layer_idx = 0
        for res in self.block_resolutions:
            in_channels = channels_dict[res] if res < img_resolution else 0
            tmp_channels = channels_dict[res]
            out_channels = channels_dict[res // 2]
            use_fp16 = (res >= fp16_resolution)
            block = DiscriminatorBlock(in_channels,
                                       tmp_channels,
                                       out_channels,
                                       resolution=res,
                                       first_layer_idx=cur_layer_idx,
                                       use_fp16=use_fp16,
                                       **block_kwargs,
                                       **common_kwargs)
            setattr(self, f'b{res}', block)
            cur_layer_idx += block.num_layers
        if c_dim > 0:
            self.mapping = MappingNetwork(z_dim=0,
                                          c_dim=c_dim,
                                          w_dim=cmap_dim,
                                          num_ws=None,
                                          w_avg_beta=None,
                                          **mapping_kwargs)
        self.b4 = DiscriminatorEpilogue(channels_dict[4],
                                        cmap_dim=cmap_dim,
                                        resolution=4,
                                        **epilogue_kwargs,
                                        **common_kwargs)
        self.register_buffer('resample_filter',
                             upfirdn2d.setup_filter([1, 3, 3, 1]))
        self.disc_c_noise = disc_c_noise

    def forward(self, img, c, update_emas=False, **block_kwargs):
        image_raw = filtered_resizing(img['image_raw'],
                                    #   size=img['image'].shape[-1],
                                      size=img['image_sr'].shape[-1],
                                      f=self.resample_filter)
        # img = torch.cat([img['image'], image_raw], 1)
        img = torch.cat([img['image_sr'], image_raw], 1)

        _ = update_emas  # unused
        x = None
        for res in self.block_resolutions:
            block = getattr(self, f'b{res}')
            x, img = block(x, img, **block_kwargs)

        cmap = None
        if self.c_dim > 0:
            if self.disc_c_noise > 0:
                c += torch.randn_like(c) * c.std(0) * self.disc_c_noise
            cmap = self.mapping(None, c)
        x = self.b4(x, img, cmap)
        return x

    def extra_repr(self):
        return f'c_dim={self.c_dim:d}, img_resolution={self.img_resolution:d}, img_channels={self.img_channels:d}'


@persistence.persistent_class
class GeoDualDiscriminator(DualDiscriminator):
    def __init__(self, c_dim, img_resolution, img_channels, architecture='resnet', channel_base=32768, channel_max=512, num_fp16_res=4, conv_clamp=256, cmap_dim=None, disc_c_noise=0, block_kwargs={}, mapping_kwargs={}, epilogue_kwargs={}, normal_condition=False):
        super().__init__(c_dim, img_resolution, img_channels, architecture, channel_base, channel_max, num_fp16_res, conv_clamp, cmap_dim, disc_c_noise, block_kwargs, mapping_kwargs, epilogue_kwargs)
        self.normal_condition = normal_condition

    def forward(self, img, c, update_emas=False, **block_kwargs):
        image= img['image']
        image_raw = filtered_resizing(img['image_raw'],
                                      size=img['image'].shape[-1],
                                      f=self.resample_filter)
        D_input_img = torch.cat([image, image_raw], 1)

        image_depth = filtered_resizing(img['image_depth'], size=img['image'].shape[-1], f=self.resample_filter)
        if self.normal_condition and 'normal' in img:
            image_normal = filtered_resizing(img['normal'], size=img['image'].shape[-1], f=self.resample_filter)
            D_input_img = torch.cat([D_input_img, image_depth, image_normal], 1)
        else:
            D_input_img = torch.cat([D_input_img, image_depth], 1)

        img = D_input_img

        _ = update_emas  # unused
        x = None
        for res in self.block_resolutions:
            block = getattr(self, f'b{res}')
            x, img = block(x, img, **block_kwargs)

        cmap = None
        if self.c_dim > 0:
            if self.disc_c_noise > 0:
                c += torch.randn_like(c) * c.std(0) * self.disc_c_noise
            cmap = self.mapping(None, c)
        x = self.b4(x, img, cmap)
        return x

#----------------------------------------------------------------------------


@persistence.persistent_class
class DummyDualDiscriminator(torch.nn.Module):
    def __init__(
            self,
            c_dim,  # Conditioning label (C) dimensionality.
            img_resolution,  # Input resolution.
            img_channels,  # Number of input color channels.
            architecture='resnet',  # Architecture: 'orig', 'skip', 'resnet'.
            channel_base=32768,  # Overall multiplier for the number of channels.
            channel_max=512,  # Maximum number of channels in any layer.
            num_fp16_res=4,  # Use FP16 for the N highest resolutions.
            conv_clamp=256,  # Clamp the output of convolution layers to +-X, None = disable clamping.
            cmap_dim=None,  # Dimensionality of mapped conditioning label, None = default.
            block_kwargs={},  # Arguments for DiscriminatorBlock.
            mapping_kwargs={},  # Arguments for MappingNetwork.
            epilogue_kwargs={},  # Arguments for DiscriminatorEpilogue.
    ):
        super().__init__()
        img_channels *= 2

        self.c_dim = c_dim
        self.img_resolution = img_resolution
        self.img_resolution_log2 = int(np.log2(img_resolution))
        self.img_channels = img_channels
        self.block_resolutions = [
            2**i for i in range(self.img_resolution_log2, 2, -1)
        ]
        channels_dict = {
            res: min(channel_base // res, channel_max)
            for res in self.block_resolutions + [4]
        }
        fp16_resolution = max(2**(self.img_resolution_log2 + 1 - num_fp16_res),
                              8)

        if cmap_dim is None:
            cmap_dim = channels_dict[4]
        if c_dim == 0:
            cmap_dim = 0

        common_kwargs = dict(img_channels=img_channels,
                             architecture=architecture,
                             conv_clamp=conv_clamp)
        cur_layer_idx = 0
        for res in self.block_resolutions:
            in_channels = channels_dict[res] if res < img_resolution else 0
            tmp_channels = channels_dict[res]
            out_channels = channels_dict[res // 2]
            use_fp16 = (res >= fp16_resolution)
            block = DiscriminatorBlock(in_channels,
                                       tmp_channels,
                                       out_channels,
                                       resolution=res,
                                       first_layer_idx=cur_layer_idx,
                                       use_fp16=use_fp16,
                                       **block_kwargs,
                                       **common_kwargs)
            setattr(self, f'b{res}', block)
            cur_layer_idx += block.num_layers
        if c_dim > 0:
            self.mapping = MappingNetwork(z_dim=0,
                                          c_dim=c_dim,
                                          w_dim=cmap_dim,
                                          num_ws=None,
                                          w_avg_beta=None,
                                          **mapping_kwargs)
        self.b4 = DiscriminatorEpilogue(channels_dict[4],
                                        cmap_dim=cmap_dim,
                                        resolution=4,
                                        **epilogue_kwargs,
                                        **common_kwargs)
        self.register_buffer('resample_filter',
                             upfirdn2d.setup_filter([1, 3, 3, 1]))

        self.raw_fade = 1

    def forward(self, img, c, update_emas=False, **block_kwargs):
        self.raw_fade = max(0, self.raw_fade - 1 / (500000 / 32))

        image_raw = filtered_resizing(img['image_raw'],
                                      size=img['image'].shape[-1],
                                      f=self.resample_filter) * self.raw_fade
        img = torch.cat([img['image'], image_raw], 1)

        _ = update_emas  # unused
        x = None
        for res in self.block_resolutions:
            block = getattr(self, f'b{res}')
            x, img = block(x, img, **block_kwargs)

        cmap = None
        if self.c_dim > 0:
            cmap = self.mapping(None, c)
        x = self.b4(x, img, cmap)
        return x

    def extra_repr(self):
        return f'c_dim={self.c_dim:d}, img_resolution={self.img_resolution:d}, img_channels={self.img_channels:d}'


#----------------------------------------------------------------------------

# panohead
# Tri-discriminator: upsampled image, super-resolved image, and segmentation mask
# V2: first concatenate imgs and seg mask, using only one conv block
@persistence.persistent_class
class MaskDualDiscriminatorV2(torch.nn.Module):
    def __init__(self,
        c_dim,                          # Conditioning label (C) dimensionality.
        img_resolution,                 # Input resolution.
        img_channels,                   # Number of input color channels.
        seg_resolution,                 # Input resolution.
        seg_channels,                   # Number of input color channels.
        architecture        = 'resnet', # Architecture: 'orig', 'skip', 'resnet'.
        channel_base        = 32768,    # Overall multiplier for the number of channels.
        channel_max         = 512,      # Maximum number of channels in any layer.
        num_fp16_res        = 4,        # Use FP16 for the N highest resolutions.
        conv_clamp          = 256,      # Clamp the output of convolution layers to +-X, None = disable clamping.
        cmap_dim            = None,     # Dimensionality of mapped conditioning label, None = default.
        disc_c_noise        = 0,        # Corrupt camera parameters with X std dev of noise before disc. pose conditioning.
        block_kwargs        = {},       # Arguments for DiscriminatorBlock.
        mapping_kwargs      = {},       # Arguments for MappingNetwork.
        epilogue_kwargs     = {},       # Arguments for DiscriminatorEpilogue.
    ):
        super().__init__()
        img_channels = img_channels * 2 + seg_channels

        self.c_dim = c_dim
        self.img_resolution = img_resolution
        self.img_resolution_log2 = int(np.log2(img_resolution))
        self.img_channels = img_channels
        self.block_resolutions = [2 ** i for i in range(self.img_resolution_log2, 2, -1)]
        channels_dict = {res: min(channel_base // res, channel_max) for res in self.block_resolutions + [4]}
        fp16_resolution = max(2 ** (self.img_resolution_log2 + 1 - num_fp16_res), 8)

        if cmap_dim is None:
            cmap_dim = channels_dict[4]
        if c_dim == 0:
            cmap_dim = 0

        common_kwargs = dict(img_channels=img_channels, architecture=architecture, conv_clamp=conv_clamp)
        cur_layer_idx = 0
        for res in self.block_resolutions:
            in_channels = channels_dict[res] if res < img_resolution else 0
            tmp_channels = channels_dict[res]
            out_channels = channels_dict[res // 2]
            use_fp16 = (res >= fp16_resolution)
            block = DiscriminatorBlock(in_channels, tmp_channels, out_channels, resolution=res,
                first_layer_idx=cur_layer_idx, use_fp16=use_fp16, **block_kwargs, **common_kwargs)
            setattr(self, f'b{res}', block)
            cur_layer_idx += block.num_layers
        if c_dim > 0:
            self.mapping = MappingNetwork(z_dim=0, c_dim=c_dim, w_dim=cmap_dim, num_ws=None, w_avg_beta=None, **mapping_kwargs)
        self.b4 = DiscriminatorEpilogue(channels_dict[4], cmap_dim=cmap_dim, resolution=4, **epilogue_kwargs, **common_kwargs)
        self.register_buffer('resample_filter', upfirdn2d.setup_filter([1,3,3,1]))
        self.disc_c_noise = disc_c_noise

    def forward(self, img, c, update_emas=False, **block_kwargs):
        image_raw = filtered_resizing(img['image_raw'], size=img['image'].shape[-1], f=self.resample_filter)
        seg = filtered_resizing(img['image_mask'], size=img['image'].shape[-1], f=self.resample_filter)
        seg = 2 * seg - 1 # normalize to [-1,1]
        img = torch.cat([img['image'], image_raw, seg], 1)

        _ = update_emas # unused
        x = None
        for res in self.block_resolutions:
            block = getattr(self, f'b{res}')
            x, img = block(x, img, **block_kwargs)

        cmap = None
        if self.c_dim > 0:
            if self.disc_c_noise > 0: c += torch.randn_like(c) * c.std(0) * self.disc_c_noise
            cmap = self.mapping(None, c)
        x = self.b4(x, img, cmap)
        return x

    def extra_repr(self):
        return ' '.join([
            f'c_dim={self.c_dim:d},',
            f'img_resolution={self.img_resolution:d}, img_channels={self.img_channels:d},',
            f'seg_resolution={self.seg_resolution:d}, seg_channels={self.seg_channels:d}'])