File size: 16,607 Bytes
3b96cb1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# Copyright (c) Open-CD. All rights reserved.
import warnings
from typing import Dict, Optional, Union

import mmcv
import mmengine.fileio as fileio
import numpy as np
from mmcv.transforms import BaseTransform
from mmcv.transforms import LoadAnnotations as MMCV_LoadAnnotations
from mmcv.transforms import LoadImageFromFile as MMCV_LoadImageFromFile

from opencd.registry import TRANSFORMS


@TRANSFORMS.register_module()
class MultiImgLoadImageFromFile(MMCV_LoadImageFromFile):
    """Load an image pair from files.

    Required Keys:

    - img_path

    Modified Keys:

    - img
    - img_shape
    - ori_shape

    """

    def __init__(self, **kwargs) -> None:
         super().__init__(**kwargs)

    def transform(self, results: dict) -> Optional[dict]:
        """Functions to load image.

        Args:
            results (dict): Result dict from
                :class:`mmengine.dataset.BaseDataset`.

        Returns:
            dict: The dict contains loaded image and meta information.
        """

        filenames = results['img_path']
        imgs = []
        try:
            for filename in filenames:
                if self.file_client_args is not None:
                    file_client = fileio.FileClient.infer_client(
                        self.file_client_args, filename)
                    img_bytes = file_client.get(filename)
                else:
                    img_bytes = fileio.get(
                        filename, backend_args=self.backend_args)
                img = mmcv.imfrombytes(
                img_bytes, flag=self.color_type, backend=self.imdecode_backend)
                if self.to_float32:
                    img = img.astype(np.float32)
                imgs.append(img)
        except Exception as e:
            if self.ignore_empty:
                return None
            else:
                raise e
        
        results['img'] = imgs
        results['img_shape'] = imgs[0].shape[:2]
        results['ori_shape'] = imgs[0].shape[:2]
        return results


@TRANSFORMS.register_module()
class MultiImgLoadAnnotations(MMCV_LoadAnnotations):
    """Load annotations for change detection provided by dataset.

    The annotation format is as the following:

    .. code-block:: python

        {
            # Filename of change detection ground truth file.
            'seg_map_path': 'a/b/c'
        }

    After this module, the annotation has been changed to the format below:

    .. code-block:: python

        {
            # in str
            'seg_fields': List
             # In uint8 type.
            'gt_seg_map': np.ndarray (H, W)
        }

    Required Keys:

    - seg_map_path (str): Path of change detection ground truth file.

    Added Keys:

    - seg_fields (List)
    - gt_seg_map (np.uint8)

    Args:
        reduce_zero_label (bool, optional): Whether reduce all label value
            by 255. Usually used for datasets where 0 is background label.
            Defaults to None.
        imdecode_backend (str): The image decoding backend type. The backend
            argument for :func:``mmcv.imfrombytes``.
            See :fun:``mmcv.imfrombytes`` for details.
            Defaults to 'pillow'.
        backend_args (dict): Arguments to instantiate a file backend.
            See https://mmengine.readthedocs.io/en/latest/api/fileio.htm
            for details. Defaults to None.
            Notes: mmcv>=2.0.0rc4, mmengine>=0.2.0 required.
    """

    def __init__(
        self,
        reduce_zero_label=None,
        backend_args=None,
        imdecode_backend='pillow',
    ) -> None:
        super().__init__(
            with_bbox=False,
            with_label=False,
            with_seg=True,
            with_keypoints=False,
            imdecode_backend=imdecode_backend,
            backend_args=backend_args)
        self.reduce_zero_label = reduce_zero_label
        if self.reduce_zero_label is not None:
            warnings.warn('`reduce_zero_label` will be deprecated, '
                          'if you would like to ignore the zero label, please '
                          'set `reduce_zero_label=True` when dataset '
                          'initialized')
        self.imdecode_backend = imdecode_backend

    def _load_seg_map(self, results: dict) -> None:
        """Private function to load semantic segmentation annotations.

        Args:
            results (dict): Result dict from :obj:``mmcv.BaseDataset``.

        Returns:
            dict: The dict contains loaded semantic segmentation annotations.
        """

        img_bytes = fileio.get(
            results['seg_map_path'], backend_args=self.backend_args)
        gt_semantic_seg = mmcv.imfrombytes(
            img_bytes, flag='grayscale', # in mmseg: unchanged
            backend=self.imdecode_backend).squeeze().astype(np.uint8)

        # reduce zero_label
        if self.reduce_zero_label is None:
            self.reduce_zero_label = results['reduce_zero_label']
        assert self.reduce_zero_label == results['reduce_zero_label'], \
            'Initialize dataset with `reduce_zero_label` as ' \
            f'{results["reduce_zero_label"]} but when load annotation ' \
            f'the `reduce_zero_label` is {self.reduce_zero_label}'
        if self.reduce_zero_label:
            # avoid using underflow conversion
            gt_semantic_seg[gt_semantic_seg == 0] = 255
            gt_semantic_seg = gt_semantic_seg - 1
            gt_semantic_seg[gt_semantic_seg == 254] = 255
        # modify to format ann
        if results.get('format_seg_map', None) is not None:
            if results['format_seg_map'] == 'to_binary':
                gt_semantic_seg_copy = gt_semantic_seg.copy()
                gt_semantic_seg[gt_semantic_seg_copy < 128] = 0
                gt_semantic_seg[gt_semantic_seg_copy >= 128] = 1
            else:
                raise ValueError('Invalid value {}'.format(results['format_seg_map']))
        # modify if custom classes
        if results.get('label_map', None) is not None:
            # Add deep copy to solve bug of repeatedly
            # replace `gt_semantic_seg`, which is reported in
            # https://github.com/open-mmlab/mmsegmentation/pull/1445/
            gt_semantic_seg_copy = gt_semantic_seg.copy()
            for old_id, new_id in results['label_map'].items():
                gt_semantic_seg[gt_semantic_seg_copy == old_id] = new_id
        results['gt_seg_map'] = gt_semantic_seg
        results['seg_fields'].append('gt_seg_map')

    def __repr__(self) -> str:
        repr_str = self.__class__.__name__
        repr_str += f'(reduce_zero_label={self.reduce_zero_label}, '
        repr_str += f"imdecode_backend='{self.imdecode_backend}', "
        repr_str += f'backend_args={self.backend_args})'
        return repr_str


@TRANSFORMS.register_module()
class MultiImgMultiAnnLoadAnnotations(MMCV_LoadAnnotations):
    """Load annotations for semantic change detection provided by dataset.

    The annotation format is as the following:

    .. code-block:: python

        {
            # Filename of change detection ground truth file.
            'seg_map_path': 'a/b/c'
        }

    After this module, the annotation has been changed to the format below:

    .. code-block:: python

        {
            # in str
            'seg_fields': List
             # In uint8 type.
            'gt_seg_map': np.ndarray (H, W)
        }

    Required Keys:

    - seg_map_path (str): Path of change detection ground truth file.

    Added Keys:

    - seg_fields (List)
    - gt_seg_map (np.uint8)

    Args:
        reduce_semantic_zero_label (bool, optional): Whether reduce all label value
            by 255. Usually used for datasets where 0 is background label.
            Defaults to None.
        imdecode_backend (str): The image decoding backend type. The backend
            argument for :func:``mmcv.imfrombytes``.
            See :fun:``mmcv.imfrombytes`` for details.
            Defaults to 'pillow'.
        backend_args (dict): Arguments to instantiate a file backend.
            See https://mmengine.readthedocs.io/en/latest/api/fileio.htm
            for details. Defaults to None.
            Notes: mmcv>=2.0.0rc4, mmengine>=0.2.0 required.
    """

    def __init__(
        self,
        reduce_semantic_zero_label=None,
        backend_args=None,
        imdecode_backend='pillow',
    ) -> None:
        super().__init__(
            with_bbox=False,
            with_label=False,
            with_seg=True,
            with_keypoints=False,
            imdecode_backend=imdecode_backend,
            backend_args=backend_args)
        self.reduce_semantic_zero_label = reduce_semantic_zero_label
        if self.reduce_semantic_zero_label is not None:
            warnings.warn('`reduce_semantic_zero_label` will be deprecated, '
                          'if you would like to ignore the zero label, please '
                          'set `reduce_semantic_zero_label=True` when dataset '
                          'initialized')
        self.imdecode_backend = imdecode_backend

    def _load_seg_map(self, results: dict) -> None:
        """Private function to load semantic segmentation annotations.

        Args:
            results (dict): Result dict from :obj:``mmcv.BaseDataset``.

        Returns:
            dict: The dict contains loaded semantic segmentation annotations.
        """

        img_bytes = fileio.get(
            results['seg_map_path'], backend_args=self.backend_args)
        gt_semantic_seg = mmcv.imfrombytes(
            img_bytes, flag='grayscale', # in mmseg: unchanged
            backend=self.imdecode_backend).squeeze().astype(np.uint8)
        # for semantic anns
        img_bytes_from = fileio.get(
            results['seg_map_path_from'], backend_args=self.backend_args)
        gt_semantic_seg_from = mmcv.imfrombytes(
            img_bytes_from, flag='grayscale',
            backend=self.imdecode_backend).squeeze().astype(np.uint8)
        img_bytes_to = fileio.get(
            results['seg_map_path_to'], backend_args=self.backend_args)
        gt_semantic_seg_to = mmcv.imfrombytes(
            img_bytes_to, flag='grayscale',
            backend=self.imdecode_backend).squeeze().astype(np.uint8)

        # reduce zero_label
        if self.reduce_semantic_zero_label is None:
            self.reduce_semantic_zero_label = results['reduce_semantic_zero_label']
        assert self.reduce_semantic_zero_label == results['reduce_semantic_zero_label'], \
            'Initialize dataset with `reduce_semantic_zero_label` as ' \
            f'{results["reduce_semantic_zero_label"]} but when load annotation ' \
            f'the `reduce_semantic_zero_label` is {self.reduce_semantic_zero_label}'
        if self.reduce_semantic_zero_label:
            # avoid using underflow conversion
            gt_semantic_seg_from[gt_semantic_seg_from == 0] = 255
            gt_semantic_seg_from = gt_semantic_seg_from - 1
            gt_semantic_seg_from[gt_semantic_seg_from == 254] = 255
            gt_semantic_seg_to[gt_semantic_seg_to == 0] = 255
            gt_semantic_seg_to = gt_semantic_seg_to - 1
            gt_semantic_seg_to[gt_semantic_seg_to == 254] = 255
        # modify to format ann
        if results.get('format_seg_map', None) is not None:
            if results['format_seg_map'] == 'to_binary':
                gt_semantic_seg_copy = gt_semantic_seg.copy()
                gt_semantic_seg[gt_semantic_seg_copy < 128] = 0
                gt_semantic_seg[gt_semantic_seg_copy >= 128] = 1
            else:
                raise ValueError('Invalid value {}'.format(results['format_seg_map']))
        # modify if custom classes
        if results.get('label_map', None) is not None:
            # Add deep copy to solve bug of repeatedly
            # replace `gt_semantic_seg`, which is reported in
            # https://github.com/open-mmlab/mmsegmentation/pull/1445/
            gt_semantic_seg_copy = gt_semantic_seg.copy()
            for old_id, new_id in results['label_map'].items():
                gt_semantic_seg[gt_semantic_seg_copy == old_id] = new_id
        if results.get('semantic_label_map', None) is not None:
            ''' Just for semantic anns here '''
            # Add deep copy to solve bug of repeatedly
            # replace `gt_semantic_seg`, which is reported in
            # https://github.com/open-mmlab/mmsegmentation/pull/1445/
            gt_semantic_seg_from_copy = gt_semantic_seg_from.copy()
            for old_id, new_id in results['label_map'].items():
                gt_semantic_seg_from[gt_semantic_seg_from_copy == old_id] = new_id
            gt_semantic_seg_to_copy = gt_semantic_seg_to.copy()
            for old_id, new_id in results['label_map'].items():
                gt_semantic_seg_to[gt_semantic_seg_to_copy == old_id] = new_id

        results['gt_seg_map'] = gt_semantic_seg
        results['gt_seg_map_from'] = gt_semantic_seg_from
        results['gt_seg_map_to'] = gt_semantic_seg_to
        results['seg_fields'].extend(['gt_seg_map', 
            'gt_seg_map_from', 'gt_seg_map_to'])

    def __repr__(self) -> str:
        repr_str = self.__class__.__name__
        repr_str += f'(reduce_semantic_zero_label={self.reduce_semantic_zero_label}, '
        repr_str += f"imdecode_backend='{self.imdecode_backend}', "
        repr_str += f'backend_args={self.backend_args})'
        return repr_str


@TRANSFORMS.register_module()
class MultiImgLoadLoadImageFromNDArray(MultiImgLoadImageFromFile):
    """Load an image pair from ``results['img']``.

    Similar with :obj:`LoadImageFromFile`, but the image has been loaded as
    :obj:`np.ndarray` in ``results['img']``. Can be used when loading image
    from webcam.

    Required Keys:

    - img

    Modified Keys:

    - img
    - img_path
    - img_shape
    - ori_shape

    Args:
        to_float32 (bool): Whether to convert the loaded image to a float32
            numpy array. If set to False, the loaded image is an uint8 array.
            Defaults to False.
    """

    def transform(self, results: dict) -> dict:
        """Transform function to add image meta information.

        Args:
            results (dict): Result dict with Webcam read image in
                ``results['img']``.

        Returns:
            dict: The dict contains loaded image and meta information.
        """

        imgs = []
        if self.to_float32:
            for img in results['img']:
                img = img.astype(np.float32)
                imgs.append(img)

        results['img_path'] = None
        results['img'] = imgs
        results['img_shape'] = imgs[0].shape[:2]
        results['ori_shape'] = imgs[0].shape[:2]
        return results


@TRANSFORMS.register_module()
class MultiImgLoadInferencerLoader(BaseTransform):
    """Load an image pair from ``results['img']``.

    Similar with :obj:`LoadImageFromFile`, but the image has been loaded as
    :obj:`np.ndarray` in ``results['img']``. Can be used when loading image
    from webcam.

    Required Keys:

    - img

    Modified Keys:

    - img
    - img_path
    - img_shape
    - ori_shape

    Args:
        to_float32 (bool): Whether to convert the loaded image to a float32
            numpy array. If set to False, the loaded image is an uint8 array.
            Defaults to False.
    """

    def __init__(self, **kwargs) -> None:
        super().__init__()
        self.from_file = TRANSFORMS.build(
            dict(type='MultiImgLoadImageFromFile', **kwargs))
        self.from_ndarray = TRANSFORMS.build(
            dict(type='MultiImgLoadLoadImageFromNDArray', **kwargs))

    def transform(self, single_input: Union[str, np.ndarray, dict]) -> dict:
        """Transform function to add image meta information.

        Args:
            results (dict): Result dict with Webcam read image in
                ``results['img']``.

        Returns:
            dict: The dict contains loaded image and meta information.
        """
        assert len(single_input) == 2, \
            'In `MultiImgLoadInferencerLoader`,' \
            '`single_input` contains bi-temporal images'
        if isinstance(single_input[0], str):
            inputs = dict(img_path=single_input)
        elif isinstance(single_input[0], Union[np.ndarray, list]):
            inputs = dict(img=single_input)
        elif isinstance(single_input[0], dict):
            inputs = single_input
        else:
            raise NotImplementedError

        if 'img' in inputs:
            return self.from_ndarray(inputs)
        return self.from_file(inputs)