语义分割常用数据集整理

发表于 | 分类于

1. Pascal VOC

21个类，1464张用于训练，1449张用于验证，测试集有1456张图片

1
0=background，1=aeroplane, 2=bicycle, 3=bird, 4=boat, 5=bottle, 6=bus, 7=car , 8=cat, 9=chair, 10=cow, 11=diningtable, 12=dog, 13=horse, 14=motorbike, 15=person, 16=potted plant, 17=sheep, 18=sofa, 19=train, 20=tv/monitor，255= 'void' or unlabelled

2. MicroSoft COCO

118287张训练图片，5000张验证图片，以及超过40670张测试图片

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
80个物体类别，分别为：['background = 0',
'person=1', 
'bicycle=2', 
'car=3', 
'motorcycle=4', 
'airplane=5', 
'bus=6', 
'train=7', 
'truck=8', 
'boat=9', 
'traffic light=10', 
'fire hydrant=11', 
'stop sign=13', 
'parking meter=14', 
'bench=15', 
'bird=16', 
'cat=17', 
'dog=18', 
'horse=19', 
'sheep=20', 
'cow=21', 
'elephant=22', 
'bear=23', 
'zebra=24', 
'giraffe=25', 
'backpack=27', 
'umbrella=28', 
'handbag=31', 
'tie=32', 
'suitcase=33', 
'frisbee=34', 
'skis=35', 
'snowboard=36', 
'sports ball=37', 
'kite=38', 
'baseball bat=39', 
'baseball glove=40', 
'skateboard=41', 
'surfboard=42', 
'tennis racket=43', 
'bottle=44', 
'wine glass=46', 
'cup=47', 
'fork=48', 
'knife=49', 
'spoon=50', 
'bowl=51', 
'banana=52', 
'apple=53', 
'sandwich=54', 
'orange=55', 
'broccoli=56', 
'carrot=57', 
'hot dog=58', 
'pizza=59', 
'donut=60', 
'cake=61', 
'chair=62', 
'couch=63', 
'potted plant=64', 
'bed=65', 
'dining table=67', 
'toilet=70', 
'tv=72', 
'laptop=73', 
'mouse=74', 
'remote=75', 
'keyboard=76', 
'cell phone=77', 
'microwave=78', 
'oven=79', 
'toaster=80', 
'sink=81', 
'refrigerator=82', 
'book=84', 
'clock=85', 
'vase=86', 
'scissors=87', 
'teddy bear=88', 
'hair drier=89', 
'toothbrush=90']。

    91个填充类别，分别为['banner=92', 'blanket=93', 'branch=94', 'bridge=95', 'building-other=96', 
    'bush=97', 'cabinet=98', 'cage=99', 'cardboard=100', 'carpet=101', 'ceiling-other=102', 
    'ceiling-tile=103', 'cloth=104', 'clothes=105', 'clouds=106', 'counter=107', 'cupboard=108', 
    'curtain=109', 'desk-stuff=110', 'dirt=111', 'door-stuff=112', 'fence=113', 'floor-marble=114', 
    'floor-other=115', 'floor-stone=116', 'floor-tile=117', 'floor-wood=118', 'flower=119', 
    'fog=120', 'food-other=121', 'fruit=122', 'furniture-other=123', 'grass=124', 'gravel=125', 
    'ground-other=126', 'hill=127', 'house=128', 'leaves=129', 'light=130', 'mat=131', 
    'metal=132', 'mirror-stuff=133', 'moss=134', 'mountain=135', 'mud=136', 'napkin=137', 
    'net=138', 'paper=139', 'pavement=140', 'pillow=141', 'plant-other=142', 'plastic=143', 
    'platform=144', 'playingfield=145', 'railing=146', 'railroad=147', 'river=148', 'road=149', 
    'rock=150', 'roof=151', 'rug=152', 'salad=153', 'sand=154', 'sea=155', 'shelf=156', 
    'sky-other=157', 'skyscraper=158', 'snow=159', 'solid-other=160', 'stairs=161', 'stone=162',
    'straw=163', 'structural-other=164', 'table=165', 'tent=166', 'textile-other=167', 
    'towel=168', 'tree=169', 'vegetable=170', 'wall-brick=171', 'wall-concrete=172', 
    'wall-other=173', 'wall-panel=174', 'wall-stone=175', 'wall-tile=176', 'wall-wood=177',
    'water-other=178', 'waterdrops=179', 'window-blind=180', 'window-other=181', 'wood=182',
    'other=183']。

除了分割外，还有其他的场景数据集

3. KITTI

训练集：200，测试集：200

4. cityscapes 街景数据集

Cityscapes包含50个欧洲城市不同场景、不同背景、不同季节的街景的33类标注物体，包括：'unlabeled'=0 , 'ego vehicle'=1 , 'rectification border'=2 , 'out of roi'= 3 , 'static'=4 , 'dynamic'=5 , 'ground'=6 ,'road'=7 ,'sidewalk'=8 ,parking'=9 ,'rail track'=10 ,'building'=11 ,'wall'=12 ,'fence'=13 , 'guard rail'=14 ,'bridge'=15 ,'tunnel'=16 ,'pole'=17 ,'polegroup'=18 , 'traffic light'=19 ,'traffic sign'=20 , 'vegetation'=21 , 'terrain'=22 ,'sky'=23 , 'person'=24 , 'rider'=25 , 'car'=26 ,'truck'=27 , 'bus'=28 ,'caravan'=29 ,'trailer'=30 ,'train'=31 ,'motorcycle'=32 , 'bicycle'=33 ，但是在这33个类中，评估时只用到了19个类别，因此训练时将33个类映射为19个类，评估时需要将19个类又映射回33个类上传评估服务器。这个数据需要注册账号才能下载。Cityscapes数据集共有fine和coarse两套评测标准，前者提供5000张精细标注的图像，后者提供5000张精细标注外加20000张粗糙标注的图像，用PASCAL VOC标准的 intersection-over-union （IoU）得分来对算法性能进行评价。 5000张精细标注的图片分为训练集2975张图片，验证集有500张图片，而测试集有1525张图片，测试集不对外公布，需要将预测结果上传到评估服务器才能计算mIoU值。

5. ADE20K

包含151个类别（包括背景），包括各种物体（比如人、汽车等）、场景（天空、路面等）

训练集由20210张场景图片组成，验证集由2000张图片构成，测试集有3352张图片组成。

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
'background' = 0
'wall=1'
'building, edifice=2'
'sky=3'
'floor, flooring=4'
'tree=5'
'ceiling=6'
'road, route=7'
'bed =8'
'windowpane, window =9'
'grass=10'
'cabinet=11'
'sidewalk, pavement=12'
'person, individual, someone, somebody, mortal, soul=13'
'earth, ground=14'
'door, double door=15'
'table=16'
'mountain, mount=17'
'plant, flora, plant life=18'
'curtain, drape, drapery, mantle, pall=19'
'chair=20'
'car, auto, automobile, machine, motorcar=21'
'water=22 '
'painting, picture=23'
'sofa, couch, lounge=24 '
'shelf=25 '
'house=26 '
'sea=27 '
'mirror=28'
'rug, carpet, carpeting=29'
'field=30'
'armchair=31'
'seat=32'
'fence, fencing=33'
'desk=34'
'rock, stone=35'
'wardrobe, closet, press=36'
'lamp=37'
'bathtub, bathing tub, bath, tub=38'
'railing, rail=39'
'cushion=40'
'base, pedestal, stand=41'
'box=42'
'column, pillar=43'
'signboard, sign=44'
'chest of drawers, chest, bureau, dresser=45'
'counter=46'
'sand=47'
'sink=48'
'skyscraper=49'
'fireplace, hearth, open fireplace=50'
'refrigerator, icebox=51'
'grandstand, covered stand=52'
'path=53'
'stairs, steps=54'
'runway=55'
'case, display case, showcase, vitrine=56'
'pool table, billiard table, snooker table=57'
'pillow=58'
'screen door, screen=59'
'stairway, staircase=60'
'river=61'
'bridge, span=62'
'bookcase=63'
'blind, screen=64'
'coffee table, cocktail table=65'
'toilet, can, commode, crapper, pot, potty, stool, throne=66'
'flower=67'
'book=68'
'hill=69'
'bench=70'
'countertop=71'
'stove, kitchen stove, range, kitchen range, cooking stove=72'
'palm, palm tree=73'
'kitchen island=74'
'computer, computing machine, computing device, data processor, electronic computer, information processing system=75'
'swivel chair=76'
'boat=77'
'bar=78'
'arcade machine=79'
'hovel, hut, hutch, shack, shanty=80'
'bus, autobus, coach, charabanc, double-decker, jitney, motorbus, motorcoach, omnibus, passenger vehicle=81'
'towel=82'
'light, light source=83'
'truck, motortruck=84'
'tower=85'
'chandelier, pendant, pendent=86'
'awning, sunshade, sunblind=87'
'streetlight, street lamp=88'
'booth, cubicle, stall, kiosk=89'
'television receiver, television, television set, tv, tv set, idiot box, boob tube, telly, goggle box=90'
'airplane, aeroplane, plane=91'
'dirt track=92'
'apparel, wearing apparel, dress, clothes=93'
'pole=94'
'land, ground, soil=95'
'bannister, banister, balustrade, balusters, handrail=96'
'escalator, moving staircase, moving stairway=97'
'ottoman, pouf, pouffe, puff, hassock=98'
'bottle=99'
'buffet, counter, sideboard=100'
'poster, posting, placard, notice, bill, card=101'
'stage=102'
'van=103'
'ship=104'
'fountain=105'
'conveyer belt, conveyor belt, conveyer, conveyor, transporter=106'
'canopy=107'
'washer, automatic washer, washing machine=108'
'plaything, toy=109'
'swimming pool, swimming bath, natatorium=110'
'stool=111'
'barrel, cask=112'
'basket, handbasket=113'
'waterfall, falls=114'
'tent, collapsible shelter=115'
'bag=116'
'minibike, motorbike=117'
'cradle=118'
'oven=119'
'ball=120'
'food, solid food=121'
'step, stair=122'
'tank, storage tank=123'
'trade name, brand name, brand, marque=124'
'microwave, microwave oven=125'
'pot, flowerpot=126'
'animal, animate being, beast, brute, creature, fauna=127'
'bicycle, bike, wheel, cycle =128'
'lake=129'
'dishwasher, dish washer, dishwashing machine=130'
'screen, silver screen, projection screen=131'
'blanket, cover=132'
'sculpture=133'
'hood, exhaust hood=134'
'sconce=135'
'vase=136'
'traffic light, traffic signal, stoplight=137'
'tray=138'
'ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, dustbin, trash barrel, trash bin=139'
'fan=140'
'pier, wharf, wharfage, dock=141'
'crt screen=142'
'plate=143'
'monitor, monitoring device=144'
'bulletin board, notice board=145'
'shower=146'
'radiator=147'
'glass, drinking glass=148'
'clock=149'
'flag=150'

6. A2D

来源：youtube上面的3782个视频，actor-action数据集（物体<->行为）

actor包括adult, baby, bird, cat and dog, as well as rigid ones, such as ball and car

8个actions： climbing, crawling, eating, flying, jumping, rolling, running, and walking

去除一些不可能的，共43个pair。

7. DAVIS

算法竞赛数据集，从2016开始，每年都有举办，2019年已经开始DAVIS2019。

DAVIS2016提供50个高质量，全高清的视频序列组，包含有多个视频目标分割挑战，如遮挡，运动模糊和外观变化。每一个视频都是稠密标注，像素级别的精度和逐帧的真值分割，构成包括50个序列总共3455标注帧，视频帧率为24fps，1080p分辨率。

DAVIS2017中又加了部分数据集。

训练集：90个小视频，给了每一帧的原图和分割图

测试集：30个小视频，给了每一帧的原图，和第一帧的分割图

比赛测试集：30个小视频，给了每一帧的原图，和第一帧的分割图

共有78个类（如下所示），每一个视频里有若干个类的物体，每个类可能有多个个体。每个视频的帧数不固定，短的30多帧，长的90多帧。

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
{
    "airplane": {
        "id": 1,
        "super_category": "vehicle"
    },
    "backpack": {
        "id": 2,
        "super_category": "accessory"
    },
    "ball": {
        "id": 3,
        "super_category": "sports"
    },
    "bear": {
        "id": 4,
        "super_category": "animal"
    },
    "bicycle": {
        "id": 5,
        "super_category": "vehicle"
    },
    "bird": {
        "id": 6,
        "super_category": "animal"
    },
    "boat": {
        "id": 7,
        "super_category": "vehicle"
    },
    "bottle": {
        "id": 8,
        "super_category": "kitchen"
    },
    "box": {
        "id": 9,
        "super_category": "device"
    },
    "bus": {
        "id": 10,
        "super_category": "vehicle"
    },
    "camel": {
        "id": 11,
        "super_category": "animal"
    },
    "car": {
        "id": 12,
        "super_category": "vehicle"
    },
    "carriage": {
        "id": 13,
        "super_category": "vehicle"
    },
    "cat": {
        "id": 14,
        "super_category": "animal"
    },
    "cellphone": {
        "id": 15,
        "super_category": "electronic"
    },
    "chamaleon": {
        "id": 16,
        "super_category": "animal"
    },
    "cow": {
        "id": 17,
        "super_category": "animal"
    },
    "deer": {
        "id": 18,
        "super_category": "animal"
    },
    "dog": {
        "id": 19,
        "super_category": "animal"
    },
    "dolphin": {
        "id": 20,
        "super_category": "animal"
    },
    "drone": {
        "id": 21,
        "super_category": "electronic"
    },
    "elephant": {
        "id": 22,
        "super_category": "animal"
    },
    "excavator": {
        "id": 23,
        "super_category": "vehicle"
    },
    "fish": {
        "id": 24,
        "super_category": "animal"
    },
    "goat": {
        "id": 25,
        "super_category": "animal"
    },
    "golf cart": {
        "id": 26,
        "super_category": "vehicle"
    },
    "golf club": {
        "id": 27,
        "super_category": "sports"
    },
    "grass": {
        "id": 28,
        "super_category": "outdoor"
    },
    "guitar": {
        "id": 29,
        "super_category": "instrument"
    },
    "gun": {
        "id": 30,
        "super_category": "sports"
    },
    "helicopter": {
        "id": 31,
        "super_category": "vehicle"
    },
    "horse": {
        "id": 32,
        "super_category": "animal"
    },
    "hoverboard": {
        "id": 33,
        "super_category": "sports"
    },
    "kart": {
        "id": 34,
        "super_category": "vehicle"
    },
    "key": {
        "id": 35,
        "super_category": "device"
    },
    "kite": {
        "id": 36,
        "super_category": "sports"
    },
    "koala": {
        "id": 37,
        "super_category": "animal"
    },
    "leash": {
        "id": 38,
        "super_category": "device"
    },
    "lion": {
        "id": 39,
        "super_category": "animal"
    },
    "lock": {
        "id": 40,
        "super_category": "device"
    },
    "mask": {
        "id": 41,
        "super_category": "accessory"
    },
    "microphone": {
        "id": 42,
        "super_category": "electronic"
    },
    "monkey": {
        "id": 43,
        "super_category": "animal"
    },
    "motorcycle": {
        "id": 44,
        "super_category": "vehicle"
    },
    "oar": {
        "id": 45,
        "super_category": "sports"
    },
    "paper": {
        "id": 46,
        "super_category": "device"
    },
    "paraglide": {
        "id": 47,
        "super_category": "sports"
    },
    "person": {
        "id": 48,
        "super_category": "person"
    },
    "pig": {
        "id": 49,
        "super_category": "animal"
    },
    "pole": {
        "id": 50,
        "super_category": "sports"
    },
    "potted plant": {
        "id": 51,
        "super_category": "furniture"
    },
    "puck": {
        "id": 52,
        "super_category": "sports"
    },
    "rack": {
        "id": 53,
        "super_category": "furniture"
    },
    "rhino": {
        "id": 54,
        "super_category": "animal"
    },
    "rope": {
        "id": 55,
        "super_category": "sports"
    },
    "sail": {
        "id": 56,
        "super_category": "sports"
    },
    "scale": {
        "id": 57,
        "super_category": "appliance"
    },
    "scooter": {
        "id": 58,
        "super_category": "vehicle"
    },
    "selfie stick": {
        "id": 59,
        "super_category": "device"
    },
    "sheep": {
        "id": 60,
        "super_category": "animal"
    },
    "skateboard": {
        "id": 61,
        "super_category": "sports"
    },
    "ski": {
        "id": 62,
        "super_category": "sports"
    },
    "ski poles": {
        "id": 63,
        "super_category": "sports"
    },
    "snake": {
        "id": 64,
        "super_category": "animal"
    },
    "snowboard": {
        "id": 65,
        "super_category": "sports"
    },
    "stick": {
        "id": 66,
        "super_category": "sports"
    },
    "stroller": {
        "id": 67,
        "super_category": "vehicle"
    },
    "surfboard": {
        "id": 68,
        "super_category": "sports"
    },
    "swing": {
        "id": 69,
        "super_category": "outdoor"
    },
    "tennis racket": {
        "id": 70,
        "super_category": "sports"
    },
    "tractor": {
        "id": 71,
        "super_category": "vehicle"
    },
    "trailer": {
        "id": 72,
        "super_category": "vehicle"
    },
    "train": {
        "id": 73,
        "super_category": "vehicle"
    },
    "truck": {
        "id": 74,
        "super_category": "vehicle"
    },
    "turtle": {
        "id": 75,
        "super_category": "animal"
    },
    "varanus": {
        "id": 76,
        "super_category": "animal"
    },
    "violin": {
        "id": 77,
        "super_category": "instrument"
    },
    "wheelchair": {
        "id": 78,
        "super_category": "vehicle"
    }
}

DAVIS2019挑战赛分为3个主题，弱监督、交互、无监督，CVPR2019的workshop，5月24日截止。

交互教程

无监督 -> coming soon

8. YouTube-VOS

共有4,453个视频组成，其中训练集（3471），验证集（474）和测试集（508），94个类

Google Drive

百度云

OneDrive