File size: 32,678 Bytes
51f6859
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
# Copyright (c) OpenMMLab. All rights reserved.
from multiprocessing import Pool

import mmcv
import numpy as np
from mmcv.utils import print_log
from terminaltables import AsciiTable

from .bbox_overlaps import bbox_overlaps
from .class_names import get_classes


def average_precision(recalls, precisions, mode='area'):
    """Calculate average precision (for single or multiple scales).

    Args:
        recalls (ndarray): shape (num_scales, num_dets) or (num_dets, )
        precisions (ndarray): shape (num_scales, num_dets) or (num_dets, )
        mode (str): 'area' or '11points', 'area' means calculating the area
            under precision-recall curve, '11points' means calculating
            the average precision of recalls at [0, 0.1, ..., 1]

    Returns:
        float or ndarray: calculated average precision
    """
    no_scale = False
    if recalls.ndim == 1:
        no_scale = True
        recalls = recalls[np.newaxis, :]
        precisions = precisions[np.newaxis, :]
    assert recalls.shape == precisions.shape and recalls.ndim == 2
    num_scales = recalls.shape[0]
    ap = np.zeros(num_scales, dtype=np.float32)
    if mode == 'area':
        zeros = np.zeros((num_scales, 1), dtype=recalls.dtype)
        ones = np.ones((num_scales, 1), dtype=recalls.dtype)
        mrec = np.hstack((zeros, recalls, ones))
        mpre = np.hstack((zeros, precisions, zeros))
        for i in range(mpre.shape[1] - 1, 0, -1):
            mpre[:, i - 1] = np.maximum(mpre[:, i - 1], mpre[:, i])
        for i in range(num_scales):
            ind = np.where(mrec[i, 1:] != mrec[i, :-1])[0]
            ap[i] = np.sum(
                (mrec[i, ind + 1] - mrec[i, ind]) * mpre[i, ind + 1])
    elif mode == '11points':
        for i in range(num_scales):
            for thr in np.arange(0, 1 + 1e-3, 0.1):
                precs = precisions[i, recalls[i, :] >= thr]
                prec = precs.max() if precs.size > 0 else 0
                ap[i] += prec
        ap /= 11
    else:
        raise ValueError(
            'Unrecognized mode, only "area" and "11points" are supported')
    if no_scale:
        ap = ap[0]
    return ap


def tpfp_imagenet(det_bboxes,
                  gt_bboxes,
                  gt_bboxes_ignore=None,
                  default_iou_thr=0.5,
                  area_ranges=None,
                  use_legacy_coordinate=False,
                  **kwargs):
    """Check if detected bboxes are true positive or false positive.

    Args:
        det_bbox (ndarray): Detected bboxes of this image, of shape (m, 5).
        gt_bboxes (ndarray): GT bboxes of this image, of shape (n, 4).
        gt_bboxes_ignore (ndarray): Ignored gt bboxes of this image,
            of shape (k, 4). Default: None
        default_iou_thr (float): IoU threshold to be considered as matched for
            medium and large bboxes (small ones have special rules).
            Default: 0.5.
        area_ranges (list[tuple] | None): Range of bbox areas to be evaluated,
            in the format [(min1, max1), (min2, max2), ...]. Default: None.
        use_legacy_coordinate (bool): Whether to use coordinate system in
            mmdet v1.x. which means width, height should be
            calculated as 'x2 - x1 + 1` and 'y2 - y1 + 1' respectively.
            Default: False.

    Returns:
        tuple[np.ndarray]: (tp, fp) whose elements are 0 and 1. The shape of
        each array is (num_scales, m).
    """

    if not use_legacy_coordinate:
        extra_length = 0.
    else:
        extra_length = 1.

    # an indicator of ignored gts
    gt_ignore_inds = np.concatenate(
        (np.zeros(gt_bboxes.shape[0],
                  dtype=bool), np.ones(gt_bboxes_ignore.shape[0], dtype=bool)))
    # stack gt_bboxes and gt_bboxes_ignore for convenience
    gt_bboxes = np.vstack((gt_bboxes, gt_bboxes_ignore))

    num_dets = det_bboxes.shape[0]
    num_gts = gt_bboxes.shape[0]
    if area_ranges is None:
        area_ranges = [(None, None)]
    num_scales = len(area_ranges)
    # tp and fp are of shape (num_scales, num_gts), each row is tp or fp
    # of a certain scale.
    tp = np.zeros((num_scales, num_dets), dtype=np.float32)
    fp = np.zeros((num_scales, num_dets), dtype=np.float32)
    if gt_bboxes.shape[0] == 0:
        if area_ranges == [(None, None)]:
            fp[...] = 1
        else:
            det_areas = (
                det_bboxes[:, 2] - det_bboxes[:, 0] + extra_length) * (
                    det_bboxes[:, 3] - det_bboxes[:, 1] + extra_length)
            for i, (min_area, max_area) in enumerate(area_ranges):
                fp[i, (det_areas >= min_area) & (det_areas < max_area)] = 1
        return tp, fp
    ious = bbox_overlaps(
        det_bboxes, gt_bboxes - 1, use_legacy_coordinate=use_legacy_coordinate)
    gt_w = gt_bboxes[:, 2] - gt_bboxes[:, 0] + extra_length
    gt_h = gt_bboxes[:, 3] - gt_bboxes[:, 1] + extra_length
    iou_thrs = np.minimum((gt_w * gt_h) / ((gt_w + 10.0) * (gt_h + 10.0)),
                          default_iou_thr)
    # sort all detections by scores in descending order
    sort_inds = np.argsort(-det_bboxes[:, -1])
    for k, (min_area, max_area) in enumerate(area_ranges):
        gt_covered = np.zeros(num_gts, dtype=bool)
        # if no area range is specified, gt_area_ignore is all False
        if min_area is None:
            gt_area_ignore = np.zeros_like(gt_ignore_inds, dtype=bool)
        else:
            gt_areas = gt_w * gt_h
            gt_area_ignore = (gt_areas < min_area) | (gt_areas >= max_area)
        for i in sort_inds:
            max_iou = -1
            matched_gt = -1
            # find best overlapped available gt
            for j in range(num_gts):
                # different from PASCAL VOC: allow finding other gts if the
                # best overlapped ones are already matched by other det bboxes
                if gt_covered[j]:
                    continue
                elif ious[i, j] >= iou_thrs[j] and ious[i, j] > max_iou:
                    max_iou = ious[i, j]
                    matched_gt = j
            # there are 4 cases for a det bbox:
            # 1. it matches a gt, tp = 1, fp = 0
            # 2. it matches an ignored gt, tp = 0, fp = 0
            # 3. it matches no gt and within area range, tp = 0, fp = 1
            # 4. it matches no gt but is beyond area range, tp = 0, fp = 0
            if matched_gt >= 0:
                gt_covered[matched_gt] = 1
                if not (gt_ignore_inds[matched_gt]
                        or gt_area_ignore[matched_gt]):
                    tp[k, i] = 1
            elif min_area is None:
                fp[k, i] = 1
            else:
                bbox = det_bboxes[i, :4]
                area = (bbox[2] - bbox[0] + extra_length) * (
                    bbox[3] - bbox[1] + extra_length)
                if area >= min_area and area < max_area:
                    fp[k, i] = 1
    return tp, fp


def tpfp_default(det_bboxes,
                 gt_bboxes,
                 gt_bboxes_ignore=None,
                 iou_thr=0.5,
                 area_ranges=None,
                 use_legacy_coordinate=False,
                 **kwargs):
    """Check if detected bboxes are true positive or false positive.

    Args:
        det_bbox (ndarray): Detected bboxes of this image, of shape (m, 5).
        gt_bboxes (ndarray): GT bboxes of this image, of shape (n, 4).
        gt_bboxes_ignore (ndarray): Ignored gt bboxes of this image,
            of shape (k, 4). Default: None
        iou_thr (float): IoU threshold to be considered as matched.
            Default: 0.5.
        area_ranges (list[tuple] | None): Range of bbox areas to be
            evaluated, in the format [(min1, max1), (min2, max2), ...].
            Default: None.
        use_legacy_coordinate (bool): Whether to use coordinate system in
            mmdet v1.x. which means width, height should be
            calculated as 'x2 - x1 + 1` and 'y2 - y1 + 1' respectively.
            Default: False.

    Returns:
        tuple[np.ndarray]: (tp, fp) whose elements are 0 and 1. The shape of
        each array is (num_scales, m).
    """

    if not use_legacy_coordinate:
        extra_length = 0.
    else:
        extra_length = 1.

    # an indicator of ignored gts
    gt_ignore_inds = np.concatenate(
        (np.zeros(gt_bboxes.shape[0],
                  dtype=bool), np.ones(gt_bboxes_ignore.shape[0], dtype=bool)))
    # stack gt_bboxes and gt_bboxes_ignore for convenience
    gt_bboxes = np.vstack((gt_bboxes, gt_bboxes_ignore))

    num_dets = det_bboxes.shape[0]
    num_gts = gt_bboxes.shape[0]
    if area_ranges is None:
        area_ranges = [(None, None)]
    num_scales = len(area_ranges)
    # tp and fp are of shape (num_scales, num_gts), each row is tp or fp of
    # a certain scale
    tp = np.zeros((num_scales, num_dets), dtype=np.float32)
    fp = np.zeros((num_scales, num_dets), dtype=np.float32)

    # if there is no gt bboxes in this image, then all det bboxes
    # within area range are false positives
    if gt_bboxes.shape[0] == 0:
        if area_ranges == [(None, None)]:
            fp[...] = 1
        else:
            det_areas = (
                det_bboxes[:, 2] - det_bboxes[:, 0] + extra_length) * (
                    det_bboxes[:, 3] - det_bboxes[:, 1] + extra_length)
            for i, (min_area, max_area) in enumerate(area_ranges):
                fp[i, (det_areas >= min_area) & (det_areas < max_area)] = 1
        return tp, fp

    ious = bbox_overlaps(
        det_bboxes, gt_bboxes, use_legacy_coordinate=use_legacy_coordinate)
    # for each det, the max iou with all gts
    ious_max = ious.max(axis=1)
    # for each det, which gt overlaps most with it
    ious_argmax = ious.argmax(axis=1)
    # sort all dets in descending order by scores
    sort_inds = np.argsort(-det_bboxes[:, -1])
    for k, (min_area, max_area) in enumerate(area_ranges):
        gt_covered = np.zeros(num_gts, dtype=bool)
        # if no area range is specified, gt_area_ignore is all False
        if min_area is None:
            gt_area_ignore = np.zeros_like(gt_ignore_inds, dtype=bool)
        else:
            gt_areas = (gt_bboxes[:, 2] - gt_bboxes[:, 0] + extra_length) * (
                gt_bboxes[:, 3] - gt_bboxes[:, 1] + extra_length)
            gt_area_ignore = (gt_areas < min_area) | (gt_areas >= max_area)
        for i in sort_inds:
            if ious_max[i] >= iou_thr:
                matched_gt = ious_argmax[i]
                if not (gt_ignore_inds[matched_gt]
                        or gt_area_ignore[matched_gt]):
                    if not gt_covered[matched_gt]:
                        gt_covered[matched_gt] = True
                        tp[k, i] = 1
                    else:
                        fp[k, i] = 1
                # otherwise ignore this detected bbox, tp = 0, fp = 0
            elif min_area is None:
                fp[k, i] = 1
            else:
                bbox = det_bboxes[i, :4]
                area = (bbox[2] - bbox[0] + extra_length) * (
                    bbox[3] - bbox[1] + extra_length)
                if area >= min_area and area < max_area:
                    fp[k, i] = 1
    return tp, fp


def tpfp_openimages(det_bboxes,
                    gt_bboxes,
                    gt_bboxes_ignore=None,
                    iou_thr=0.5,
                    area_ranges=None,
                    use_legacy_coordinate=False,
                    gt_bboxes_group_of=None,
                    use_group_of=True,
                    ioa_thr=0.5,
                    **kwargs):
    """Check if detected bboxes are true positive or false positive.

    Args:
        det_bbox (ndarray): Detected bboxes of this image, of shape (m, 5).
        gt_bboxes (ndarray): GT bboxes of this image, of shape (n, 4).
        gt_bboxes_ignore (ndarray): Ignored gt bboxes of this image,
            of shape (k, 4). Default: None
        iou_thr (float): IoU threshold to be considered as matched.
            Default: 0.5.
        area_ranges (list[tuple] | None): Range of bbox areas to be
            evaluated, in the format [(min1, max1), (min2, max2), ...].
            Default: None.
        use_legacy_coordinate (bool): Whether to use coordinate system in
            mmdet v1.x. which means width, height should be
            calculated as 'x2 - x1 + 1` and 'y2 - y1 + 1' respectively.
            Default: False.
        gt_bboxes_group_of (ndarray): GT group_of of this image, of shape
            (k, 1). Default: None
        use_group_of (bool): Whether to use group of when calculate TP and FP,
            which only used in OpenImages evaluation. Default: True.
        ioa_thr (float | None): IoA threshold to be considered as matched,
            which only used in OpenImages evaluation. Default: 0.5.

    Returns:
        tuple[np.ndarray]: Returns a tuple (tp, fp, det_bboxes), where
        (tp, fp) whose elements are 0 and 1. The shape of each array is
        (num_scales, m). (det_bboxes) whose will filter those are not
        matched by group of gts when processing Open Images evaluation.
        The shape is (num_scales, m).
    """

    if not use_legacy_coordinate:
        extra_length = 0.
    else:
        extra_length = 1.

    # an indicator of ignored gts
    gt_ignore_inds = np.concatenate(
        (np.zeros(gt_bboxes.shape[0],
                  dtype=bool), np.ones(gt_bboxes_ignore.shape[0], dtype=bool)))
    # stack gt_bboxes and gt_bboxes_ignore for convenience
    gt_bboxes = np.vstack((gt_bboxes, gt_bboxes_ignore))

    num_dets = det_bboxes.shape[0]
    num_gts = gt_bboxes.shape[0]
    if area_ranges is None:
        area_ranges = [(None, None)]
    num_scales = len(area_ranges)
    # tp and fp are of shape (num_scales, num_gts), each row is tp or fp of
    # a certain scale
    tp = np.zeros((num_scales, num_dets), dtype=np.float32)
    fp = np.zeros((num_scales, num_dets), dtype=np.float32)

    # if there is no gt bboxes in this image, then all det bboxes
    # within area range are false positives
    if gt_bboxes.shape[0] == 0:
        if area_ranges == [(None, None)]:
            fp[...] = 1
        else:
            det_areas = (
                det_bboxes[:, 2] - det_bboxes[:, 0] + extra_length) * (
                    det_bboxes[:, 3] - det_bboxes[:, 1] + extra_length)
            for i, (min_area, max_area) in enumerate(area_ranges):
                fp[i, (det_areas >= min_area) & (det_areas < max_area)] = 1
        return tp, fp, det_bboxes

    if gt_bboxes_group_of is not None and use_group_of:
        # if handle group-of boxes, divided gt boxes into two parts:
        # non-group-of and group-of.Then calculate ious and ioas through
        # non-group-of group-of gts respectively. This only used in
        # OpenImages evaluation.
        assert gt_bboxes_group_of.shape[0] == gt_bboxes.shape[0]
        non_group_gt_bboxes = gt_bboxes[~gt_bboxes_group_of]
        group_gt_bboxes = gt_bboxes[gt_bboxes_group_of]
        num_gts_group = group_gt_bboxes.shape[0]
        ious = bbox_overlaps(det_bboxes, non_group_gt_bboxes)
        ioas = bbox_overlaps(det_bboxes, group_gt_bboxes, mode='iof')
    else:
        # if not consider group-of boxes, only calculate ious through gt boxes
        ious = bbox_overlaps(
            det_bboxes, gt_bboxes, use_legacy_coordinate=use_legacy_coordinate)
        ioas = None

    if ious.shape[1] > 0:
        # for each det, the max iou with all gts
        ious_max = ious.max(axis=1)
        # for each det, which gt overlaps most with it
        ious_argmax = ious.argmax(axis=1)
        # sort all dets in descending order by scores
        sort_inds = np.argsort(-det_bboxes[:, -1])
        for k, (min_area, max_area) in enumerate(area_ranges):
            gt_covered = np.zeros(num_gts, dtype=bool)
            # if no area range is specified, gt_area_ignore is all False
            if min_area is None:
                gt_area_ignore = np.zeros_like(gt_ignore_inds, dtype=bool)
            else:
                gt_areas = (
                    gt_bboxes[:, 2] - gt_bboxes[:, 0] + extra_length) * (
                        gt_bboxes[:, 3] - gt_bboxes[:, 1] + extra_length)
                gt_area_ignore = (gt_areas < min_area) | (gt_areas >= max_area)
            for i in sort_inds:
                if ious_max[i] >= iou_thr:
                    matched_gt = ious_argmax[i]
                    if not (gt_ignore_inds[matched_gt]
                            or gt_area_ignore[matched_gt]):
                        if not gt_covered[matched_gt]:
                            gt_covered[matched_gt] = True
                            tp[k, i] = 1
                        else:
                            fp[k, i] = 1
                    # otherwise ignore this detected bbox, tp = 0, fp = 0
                elif min_area is None:
                    fp[k, i] = 1
                else:
                    bbox = det_bboxes[i, :4]
                    area = (bbox[2] - bbox[0] + extra_length) * (
                        bbox[3] - bbox[1] + extra_length)
                    if area >= min_area and area < max_area:
                        fp[k, i] = 1
    else:
        # if there is no no-group-of gt bboxes in this image,
        # then all det bboxes within area range are false positives.
        # Only used in OpenImages evaluation.
        if area_ranges == [(None, None)]:
            fp[...] = 1
        else:
            det_areas = (
                det_bboxes[:, 2] - det_bboxes[:, 0] + extra_length) * (
                    det_bboxes[:, 3] - det_bboxes[:, 1] + extra_length)
            for i, (min_area, max_area) in enumerate(area_ranges):
                fp[i, (det_areas >= min_area) & (det_areas < max_area)] = 1

    if ioas is None or ioas.shape[1] <= 0:
        return tp, fp, det_bboxes
    else:
        # The evaluation of group-of TP and FP are done in two stages:
        # 1. All detections are first matched to non group-of boxes; true
        #    positives are determined.
        # 2. Detections that are determined as false positives are matched
        #    against group-of boxes and calculated group-of TP and FP.
        # Only used in OpenImages evaluation.
        det_bboxes_group = np.zeros(
            (num_scales, ioas.shape[1], det_bboxes.shape[1]), dtype=float)
        match_group_of = np.zeros((num_scales, num_dets), dtype=bool)
        tp_group = np.zeros((num_scales, num_gts_group), dtype=np.float32)
        ioas_max = ioas.max(axis=1)
        # for each det, which gt overlaps most with it
        ioas_argmax = ioas.argmax(axis=1)
        # sort all dets in descending order by scores
        sort_inds = np.argsort(-det_bboxes[:, -1])
        for k, (min_area, max_area) in enumerate(area_ranges):
            box_is_covered = tp[k]
            # if no area range is specified, gt_area_ignore is all False
            if min_area is None:
                gt_area_ignore = np.zeros_like(gt_ignore_inds, dtype=bool)
            else:
                gt_areas = (gt_bboxes[:, 2] - gt_bboxes[:, 0]) * (
                    gt_bboxes[:, 3] - gt_bboxes[:, 1])
                gt_area_ignore = (gt_areas < min_area) | (gt_areas >= max_area)
            for i in sort_inds:
                matched_gt = ioas_argmax[i]
                if not box_is_covered[i]:
                    if ioas_max[i] >= ioa_thr:
                        if not (gt_ignore_inds[matched_gt]
                                or gt_area_ignore[matched_gt]):
                            if not tp_group[k, matched_gt]:
                                tp_group[k, matched_gt] = 1
                                match_group_of[k, i] = True
                            else:
                                match_group_of[k, i] = True

                            if det_bboxes_group[k, matched_gt, -1] < \
                                    det_bboxes[i, -1]:
                                det_bboxes_group[k, matched_gt] = \
                                    det_bboxes[i]

        fp_group = (tp_group <= 0).astype(float)
        tps = []
        fps = []
        # concatenate tp, fp, and det-boxes which not matched group of
        # gt boxes and tp_group, fp_group, and det_bboxes_group which
        # matched group of boxes respectively.
        for i in range(num_scales):
            tps.append(
                np.concatenate((tp[i][~match_group_of[i]], tp_group[i])))
            fps.append(
                np.concatenate((fp[i][~match_group_of[i]], fp_group[i])))
            det_bboxes = np.concatenate(
                (det_bboxes[~match_group_of[i]], det_bboxes_group[i]))

        tp = np.vstack(tps)
        fp = np.vstack(fps)
        return tp, fp, det_bboxes


def get_cls_results(det_results, annotations, class_id):
    """Get det results and gt information of a certain class.

    Args:
        det_results (list[list]): Same as `eval_map()`.
        annotations (list[dict]): Same as `eval_map()`.
        class_id (int): ID of a specific class.

    Returns:
        tuple[list[np.ndarray]]: detected bboxes, gt bboxes, ignored gt bboxes
    """
    cls_dets = [img_res[class_id] for img_res in det_results]
    cls_gts = []
    cls_gts_ignore = []
    for ann in annotations:
        gt_inds = ann['labels'] == class_id
        cls_gts.append(ann['bboxes'][gt_inds, :])

        if ann.get('labels_ignore', None) is not None:
            ignore_inds = ann['labels_ignore'] == class_id
            cls_gts_ignore.append(ann['bboxes_ignore'][ignore_inds, :])
        else:
            cls_gts_ignore.append(np.empty((0, 4), dtype=np.float32))

    return cls_dets, cls_gts, cls_gts_ignore


def get_cls_group_ofs(annotations, class_id):
    """Get `gt_group_of` of a certain class, which is used in Open Images.

    Args:
        annotations (list[dict]): Same as `eval_map()`.
        class_id (int): ID of a specific class.

    Returns:
        list[np.ndarray]: `gt_group_of` of a certain class.
    """
    gt_group_ofs = []
    for ann in annotations:
        gt_inds = ann['labels'] == class_id
        if ann.get('gt_is_group_ofs', None) is not None:
            gt_group_ofs.append(ann['gt_is_group_ofs'][gt_inds])
        else:
            gt_group_ofs.append(np.empty((0, 1), dtype=bool))

    return gt_group_ofs


def eval_map(det_results,
             annotations,
             scale_ranges=None,
             iou_thr=0.5,
             ioa_thr=None,
             dataset=None,
             logger=None,
             tpfp_fn=None,
             nproc=4,
             use_legacy_coordinate=False,
             use_group_of=False):
    """Evaluate mAP of a dataset.

    Args:
        det_results (list[list]): [[cls1_det, cls2_det, ...], ...].
            The outer list indicates images, and the inner list indicates
            per-class detected bboxes.
        annotations (list[dict]): Ground truth annotations where each item of
            the list indicates an image. Keys of annotations are:

            - `bboxes`: numpy array of shape (n, 4)
            - `labels`: numpy array of shape (n, )
            - `bboxes_ignore` (optional): numpy array of shape (k, 4)
            - `labels_ignore` (optional): numpy array of shape (k, )
        scale_ranges (list[tuple] | None): Range of scales to be evaluated,
            in the format [(min1, max1), (min2, max2), ...]. A range of
            (32, 64) means the area range between (32**2, 64**2).
            Default: None.
        iou_thr (float): IoU threshold to be considered as matched.
            Default: 0.5.
        ioa_thr (float | None): IoA threshold to be considered as matched,
            which only used in OpenImages evaluation. Default: None.
        dataset (list[str] | str | None): Dataset name or dataset classes,
            there are minor differences in metrics for different datasets, e.g.
            "voc07", "imagenet_det", etc. Default: None.
        logger (logging.Logger | str | None): The way to print the mAP
            summary. See `mmcv.utils.print_log()` for details. Default: None.
        tpfp_fn (callable | None): The function used to determine true/
            false positives. If None, :func:`tpfp_default` is used as default
            unless dataset is 'det' or 'vid' (:func:`tpfp_imagenet` in this
            case). If it is given as a function, then this function is used
            to evaluate tp & fp. Default None.
        nproc (int): Processes used for computing TP and FP.
            Default: 4.
        use_legacy_coordinate (bool): Whether to use coordinate system in
            mmdet v1.x. which means width, height should be
            calculated as 'x2 - x1 + 1` and 'y2 - y1 + 1' respectively.
            Default: False.
        use_group_of (bool): Whether to use group of when calculate TP and FP,
            which only used in OpenImages evaluation. Default: False.

    Returns:
        tuple: (mAP, [dict, dict, ...])
    """
    assert len(det_results) == len(annotations)
    if not use_legacy_coordinate:
        extra_length = 0.
    else:
        extra_length = 1.

    num_imgs = len(det_results)
    num_scales = len(scale_ranges) if scale_ranges is not None else 1
    num_classes = len(det_results[0])  # positive class num
    area_ranges = ([(rg[0]**2, rg[1]**2) for rg in scale_ranges]
                   if scale_ranges is not None else None)

    # There is no need to use multi processes to process
    # when num_imgs = 1 .
    if num_imgs > 1:
        assert nproc > 0, 'nproc must be at least one.'
        nproc = min(nproc, num_imgs)
        pool = Pool(nproc)

    eval_results = []
    for i in range(num_classes):
        # get gt and det bboxes of this class
        cls_dets, cls_gts, cls_gts_ignore = get_cls_results(
            det_results, annotations, i)
        # choose proper function according to datasets to compute tp and fp
        if tpfp_fn is None:
            if dataset in ['det', 'vid']:
                tpfp_fn = tpfp_imagenet
            elif dataset in ['oid_challenge', 'oid_v6'] \
                    or use_group_of is True:
                tpfp_fn = tpfp_openimages
            else:
                tpfp_fn = tpfp_default
        if not callable(tpfp_fn):
            raise ValueError(
                f'tpfp_fn has to be a function or None, but got {tpfp_fn}')

        if num_imgs > 1:
            # compute tp and fp for each image with multiple processes
            args = []
            if use_group_of:
                # used in Open Images Dataset evaluation
                gt_group_ofs = get_cls_group_ofs(annotations, i)
                args.append(gt_group_ofs)
                args.append([use_group_of for _ in range(num_imgs)])
            if ioa_thr is not None:
                args.append([ioa_thr for _ in range(num_imgs)])

            tpfp = pool.starmap(
                tpfp_fn,
                zip(cls_dets, cls_gts, cls_gts_ignore,
                    [iou_thr for _ in range(num_imgs)],
                    [area_ranges for _ in range(num_imgs)],
                    [use_legacy_coordinate for _ in range(num_imgs)], *args))
        else:
            tpfp = tpfp_fn(
                cls_dets[0],
                cls_gts[0],
                cls_gts_ignore[0],
                iou_thr,
                area_ranges,
                use_legacy_coordinate,
                gt_bboxes_group_of=(get_cls_group_ofs(annotations, i)[0]
                                    if use_group_of else None),
                use_group_of=use_group_of,
                ioa_thr=ioa_thr)
            tpfp = [tpfp]

        if use_group_of:
            tp, fp, cls_dets = tuple(zip(*tpfp))
        else:
            tp, fp = tuple(zip(*tpfp))
        # calculate gt number of each scale
        # ignored gts or gts beyond the specific scale are not counted
        num_gts = np.zeros(num_scales, dtype=int)
        for j, bbox in enumerate(cls_gts):
            if area_ranges is None:
                num_gts[0] += bbox.shape[0]
            else:
                gt_areas = (bbox[:, 2] - bbox[:, 0] + extra_length) * (
                    bbox[:, 3] - bbox[:, 1] + extra_length)
                for k, (min_area, max_area) in enumerate(area_ranges):
                    num_gts[k] += np.sum((gt_areas >= min_area)
                                         & (gt_areas < max_area))
        # sort all det bboxes by score, also sort tp and fp
        cls_dets = np.vstack(cls_dets)
        num_dets = cls_dets.shape[0]
        sort_inds = np.argsort(-cls_dets[:, -1])
        tp = np.hstack(tp)[:, sort_inds]
        fp = np.hstack(fp)[:, sort_inds]
        # calculate recall and precision with tp and fp
        tp = np.cumsum(tp, axis=1)
        fp = np.cumsum(fp, axis=1)
        eps = np.finfo(np.float32).eps
        recalls = tp / np.maximum(num_gts[:, np.newaxis], eps)
        precisions = tp / np.maximum((tp + fp), eps)
        # calculate AP
        if scale_ranges is None:
            recalls = recalls[0, :]
            precisions = precisions[0, :]
            num_gts = num_gts.item()
        mode = 'area' if dataset != 'voc07' else '11points'
        ap = average_precision(recalls, precisions, mode)
        eval_results.append({
            'num_gts': num_gts,
            'num_dets': num_dets,
            'recall': recalls,
            'precision': precisions,
            'ap': ap
        })

    if num_imgs > 1:
        pool.close()

    if scale_ranges is not None:
        # shape (num_classes, num_scales)
        all_ap = np.vstack([cls_result['ap'] for cls_result in eval_results])
        all_num_gts = np.vstack(
            [cls_result['num_gts'] for cls_result in eval_results])
        mean_ap = []
        for i in range(num_scales):
            if np.any(all_num_gts[:, i] > 0):
                mean_ap.append(all_ap[all_num_gts[:, i] > 0, i].mean())
            else:
                mean_ap.append(0.0)
    else:
        aps = []
        for cls_result in eval_results:
            if cls_result['num_gts'] > 0:
                aps.append(cls_result['ap'])
        mean_ap = np.array(aps).mean().item() if aps else 0.0

    print_map_summary(
        mean_ap, eval_results, dataset, area_ranges, logger=logger)

    return mean_ap, eval_results


def print_map_summary(mean_ap,
                      results,
                      dataset=None,
                      scale_ranges=None,
                      logger=None):
    """Print mAP and results of each class.

    A table will be printed to show the gts/dets/recall/AP of each class and
    the mAP.

    Args:
        mean_ap (float): Calculated from `eval_map()`.
        results (list[dict]): Calculated from `eval_map()`.
        dataset (list[str] | str | None): Dataset name or dataset classes.
        scale_ranges (list[tuple] | None): Range of scales to be evaluated.
        logger (logging.Logger | str | None): The way to print the mAP
            summary. See `mmcv.utils.print_log()` for details. Default: None.
    """

    if logger == 'silent':
        return

    if isinstance(results[0]['ap'], np.ndarray):
        num_scales = len(results[0]['ap'])
    else:
        num_scales = 1

    if scale_ranges is not None:
        assert len(scale_ranges) == num_scales

    num_classes = len(results)

    recalls = np.zeros((num_scales, num_classes), dtype=np.float32)
    aps = np.zeros((num_scales, num_classes), dtype=np.float32)
    num_gts = np.zeros((num_scales, num_classes), dtype=int)
    for i, cls_result in enumerate(results):
        if cls_result['recall'].size > 0:
            recalls[:, i] = np.array(cls_result['recall'], ndmin=2)[:, -1]
        aps[:, i] = cls_result['ap']
        num_gts[:, i] = cls_result['num_gts']

    if dataset is None:
        label_names = [str(i) for i in range(num_classes)]
    elif mmcv.is_str(dataset):
        label_names = get_classes(dataset)
    else:
        label_names = dataset

    if not isinstance(mean_ap, list):
        mean_ap = [mean_ap]

    header = ['class', 'gts', 'dets', 'recall', 'ap']
    for i in range(num_scales):
        if scale_ranges is not None:
            print_log(f'Scale range {scale_ranges[i]}', logger=logger)
        table_data = [header]
        for j in range(num_classes):
            row_data = [
                label_names[j], num_gts[i, j], results[j]['num_dets'],
                f'{recalls[i, j]:.3f}', f'{aps[i, j]:.3f}'
            ]
            table_data.append(row_data)
        table_data.append(['mAP', '', '', '', f'{mean_ap[i]:.3f}'])
        table = AsciiTable(table_data)
        table.inner_footing_row_border = True
        print_log('\n' + table.table, logger=logger)