forked from huggingface/diffusers
-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy pathimage_processor.py
More file actions
359 lines (304 loc) 路 13 KB
/
Copy pathimage_processor.py
File metadata and controls
359 lines (304 loc) 路 13 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
# Copyright 2023 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import warnings
from typing import List, Optional, Union
import numpy as np
import PIL
import torch
from PIL import Image
from .configuration_utils import ConfigMixin, register_to_config
from .utils import CONFIG_NAME, PIL_INTERPOLATION, deprecate
class VaeImageProcessor(ConfigMixin):
"""
Image Processor for VAE
Args:
do_resize (`bool`, *optional*, defaults to `True`):
Whether to downscale the image's (height, width) dimensions to multiples of `vae_scale_factor`. Can accept
`height` and `width` arguments from `preprocess` method
vae_scale_factor (`int`, *optional*, defaults to `8`):
VAE scale factor. If `do_resize` is True, the image will be automatically resized to multiples of this
factor.
resample (`str`, *optional*, defaults to `lanczos`):
Resampling filter to use when resizing the image.
do_normalize (`bool`, *optional*, defaults to `True`):
Whether to normalize the image to [-1,1]
do_convert_rgb (`bool`, *optional*, defaults to be `False`):
Whether to convert the images to RGB format.
"""
config_name = CONFIG_NAME
@register_to_config
def __init__(
self,
do_resize: bool = True,
vae_scale_factor: int = 8,
resample: str = "lanczos",
do_normalize: bool = True,
do_convert_rgb: bool = False,
):
super().__init__()
@staticmethod
def numpy_to_pil(images: np.ndarray) -> PIL.Image.Image:
"""
Convert a numpy image or a batch of images to a PIL image.
"""
if images.ndim == 3:
images = images[None, ...]
images = (images * 255).round().astype("uint8")
if images.shape[-1] == 1:
# special case for grayscale (single channel) images
pil_images = [Image.fromarray(image.squeeze(), mode="L") for image in images]
else:
pil_images = [Image.fromarray(image) for image in images]
return pil_images
@staticmethod
def pil_to_numpy(images: Union[List[PIL.Image.Image], PIL.Image.Image]) -> np.ndarray:
"""
Convert a PIL image or a list of PIL images to numpy arrays.
"""
if not isinstance(images, list):
images = [images]
images = [np.array(image).astype(np.float32) / 255.0 for image in images]
images = np.stack(images, axis=0)
return images
@staticmethod
def numpy_to_pt(images: np.ndarray) -> torch.FloatTensor:
"""
Convert a numpy image to a pytorch tensor
"""
if images.ndim == 3:
images = images[..., None]
images = torch.from_numpy(images.transpose(0, 3, 1, 2))
return images
@staticmethod
def pt_to_numpy(images: torch.FloatTensor) -> np.ndarray:
"""
Convert a pytorch tensor to a numpy image
"""
images = images.cpu().permute(0, 2, 3, 1).float().numpy()
return images
@staticmethod
def normalize(images):
"""
Normalize an image array to [-1,1]
"""
return 2.0 * images - 1.0
@staticmethod
def denormalize(images):
"""
Denormalize an image array to [0,1]
"""
return (images / 2 + 0.5).clamp(0, 1)
@staticmethod
def convert_to_rgb(image: PIL.Image.Image) -> PIL.Image.Image:
"""
Converts an image to RGB format.
"""
image = image.convert("RGB")
return image
def resize(
self,
image: PIL.Image.Image,
height: Optional[int] = None,
width: Optional[int] = None,
) -> PIL.Image.Image:
"""
Resize a PIL image. Both height and width will be downscaled to the next integer multiple of `vae_scale_factor`
"""
if height is None:
height = image.height
if width is None:
width = image.width
width, height = (
x - x % self.config.vae_scale_factor for x in (width, height)
) # resize to integer multiple of vae_scale_factor
image = image.resize((width, height), resample=PIL_INTERPOLATION[self.config.resample])
return image
def preprocess(
self,
image: Union[torch.FloatTensor, PIL.Image.Image, np.ndarray],
height: Optional[int] = None,
width: Optional[int] = None,
) -> torch.Tensor:
"""
Preprocess the image input, accepted formats are PIL images, numpy arrays or pytorch tensors"
"""
supported_formats = (PIL.Image.Image, np.ndarray, torch.Tensor)
if isinstance(image, supported_formats):
image = [image]
elif not (isinstance(image, list) and all(isinstance(i, supported_formats) for i in image)):
raise ValueError(
f"Input is in incorrect format: {[type(i) for i in image]}. Currently, we only support {', '.join(supported_formats)}"
)
if isinstance(image[0], PIL.Image.Image):
if self.config.do_convert_rgb:
image = [self.convert_to_rgb(i) for i in image]
if self.config.do_resize:
image = [self.resize(i, height, width) for i in image]
image = self.pil_to_numpy(image) # to np
image = self.numpy_to_pt(image) # to pt
elif isinstance(image[0], np.ndarray):
image = np.concatenate(image, axis=0) if image[0].ndim == 4 else np.stack(image, axis=0)
image = self.numpy_to_pt(image)
_, _, height, width = image.shape
if self.config.do_resize and (
height % self.config.vae_scale_factor != 0 or width % self.config.vae_scale_factor != 0
):
raise ValueError(
f"Currently we only support resizing for PIL image - please resize your numpy array to be divisible by {self.config.vae_scale_factor}"
f"currently the sizes are {height} and {width}. You can also pass a PIL image instead to use resize option in VAEImageProcessor"
)
elif isinstance(image[0], torch.Tensor):
image = torch.cat(image, axis=0) if image[0].ndim == 4 else torch.stack(image, axis=0)
_, channel, height, width = image.shape
# don't need any preprocess if the image is latents
if channel == 4:
return image
if self.config.do_resize and (
height % self.config.vae_scale_factor != 0 or width % self.config.vae_scale_factor != 0
):
raise ValueError(
f"Currently we only support resizing for PIL image - please resize your pytorch tensor to be divisible by {self.config.vae_scale_factor}"
f"currently the sizes are {height} and {width}. You can also pass a PIL image instead to use resize option in VAEImageProcessor"
)
# expected range [0,1], normalize to [-1,1]
do_normalize = self.config.do_normalize
if image.min() < 0:
warnings.warn(
"Passing `image` as torch tensor with value range in [-1,1] is deprecated. The expected value range for image tensor is [0,1] "
f"when passing as pytorch tensor or numpy Array. You passed `image` with value range [{image.min()},{image.max()}]",
FutureWarning,
)
do_normalize = False
if do_normalize:
image = self.normalize(image)
return image
def postprocess(
self,
image: torch.FloatTensor,
output_type: str = "pil",
do_denormalize: Optional[List[bool]] = None,
):
if not isinstance(image, torch.Tensor):
raise ValueError(
f"Input for postprocessing is in incorrect format: {type(image)}. We only support pytorch tensor"
)
if output_type not in ["latent", "pt", "np", "pil"]:
deprecation_message = (
f"the output_type {output_type} is outdated and has been set to `np`. Please make sure to set it to one of these instead: "
"`pil`, `np`, `pt`, `latent`"
)
deprecate("Unsupported output_type", "1.0.0", deprecation_message, standard_warn=False)
output_type = "np"
if output_type == "latent":
return image
if do_denormalize is None:
do_denormalize = [self.config.do_normalize] * image.shape[0]
image = torch.stack(
[self.denormalize(image[i]) if do_denormalize[i] else image[i] for i in range(image.shape[0])]
)
if output_type == "pt":
return image
image = self.pt_to_numpy(image)
if output_type == "np":
return image
if output_type == "pil":
return self.numpy_to_pil(image)
class VaeImageProcessorLDM3D(VaeImageProcessor):
"""
Image Processor for VAE LDM3D.
Args:
do_resize (`bool`, *optional*, defaults to `True`):
Whether to downscale the image's (height, width) dimensions to multiples of `vae_scale_factor`.
vae_scale_factor (`int`, *optional*, defaults to `8`):
VAE scale factor. If `do_resize` is True, the image will be automatically resized to multiples of this
factor.
resample (`str`, *optional*, defaults to `lanczos`):
Resampling filter to use when resizing the image.
do_normalize (`bool`, *optional*, defaults to `True`):
Whether to normalize the image to [-1,1]
"""
config_name = CONFIG_NAME
@register_to_config
def __init__(
self,
do_resize: bool = True,
vae_scale_factor: int = 8,
resample: str = "lanczos",
do_normalize: bool = True,
):
super().__init__()
@staticmethod
def numpy_to_pil(images):
"""
Convert a numpy image or a batch of images to a PIL image.
"""
if images.ndim == 3:
images = images[None, ...]
images = (images * 255).round().astype("uint8")
if images.shape[-1] == 1:
# special case for grayscale (single channel) images
pil_images = [Image.fromarray(image.squeeze(), mode="L") for image in images]
else:
pil_images = [Image.fromarray(image[:, :, :3]) for image in images]
return pil_images
@staticmethod
def rgblike_to_depthmap(image):
"""
Args:
image: RGB-like depth image
Returns: depth map
"""
return image[:, :, 1] * 2**8 + image[:, :, 2]
def numpy_to_depth(self, images):
"""
Convert a numpy depth image or a batch of images to a PIL image.
"""
if images.ndim == 3:
images = images[None, ...]
images = (images * 255).round().astype("uint8")
if images.shape[-1] == 1:
# special case for grayscale (single channel) images
raise Exception("Not supported")
else:
pil_images = [Image.fromarray(self.rgblike_to_depthmap(image[:, :, 3:]), mode="I;16") for image in images]
return pil_images
def postprocess(
self,
image: torch.FloatTensor,
output_type: str = "pil",
do_denormalize: Optional[List[bool]] = None,
):
if not isinstance(image, torch.Tensor):
raise ValueError(
f"Input for postprocessing is in incorrect format: {type(image)}. We only support pytorch tensor"
)
if output_type not in ["latent", "pt", "np", "pil"]:
deprecation_message = (
f"the output_type {output_type} is outdated and has been set to `np`. Please make sure to set it to one of these instead: "
"`pil`, `np`, `pt`, `latent`"
)
deprecate("Unsupported output_type", "1.0.0", deprecation_message, standard_warn=False)
output_type = "np"
if do_denormalize is None:
do_denormalize = [self.config.do_normalize] * image.shape[0]
image = torch.stack(
[self.denormalize(image[i]) if do_denormalize[i] else image[i] for i in range(image.shape[0])]
)
image = self.pt_to_numpy(image)
if output_type == "np":
return image[:, :, :, :3], np.stack([self.rgblike_to_depthmap(im[:, :, 3:]) for im in image], axis=0)
if output_type == "pil":
return self.numpy_to_pil(image), self.numpy_to_depth(image)
else:
raise Exception(f"This type {output_type} is not supported")