README.md

metadata

license: apache-2.0
task_categories:
  - image-segmentation
tags:
  - semantic-segmentation
  - coco
size_categories:
  - n<1K
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: image
      dtype: image
    - name: timestamp_ms
      dtype: int64
    - name: width
      dtype: int64
    - name: height
      dtype: int64
    - name: machine_id
      dtype: string
    - name: image_id
      dtype: string
    - name: instance_mask
      dtype: image
    - name: semantic_mask
      dtype: image
    - name: review_status
      dtype: string
  splits:
    - name: train
      num_bytes: 19867701
      num_examples: 224
  download_size: 19864561
  dataset_size: 19867701

ndimensions/semantic-segmentation

Semantic segmentation dataset in COCO format with direct index masks.

Classes (117 total)

ID	Name	Color (RGB)
0	background	(0, 0, 0)
1	cabinet	(57, 107, 229)
2	chair	(135, 210, 31)
3	monitor	(191, 66, 175)
4	desk	(57, 229, 200)
5	curtain	(210, 128, 31)
6	door	(97, 66, 191)
7	vent	(64, 229, 57)
8	light	(210, 31, 91)
9	trash can	(66, 144, 191)
10	bed	(215, 229, 57)
11	mop	(172, 31, 210)
12	fire extinguisher	(66, 191, 128)
13	mattress	(229, 92, 57)
14	furniture	(31, 46, 210)
15	roll	(113, 191, 66)
16	box	(229, 57, 172)
17	outlet	(31, 203, 210)
18	refrigerator	(191, 159, 66)
19	painting	(135, 57, 229)
20	oven	(31, 210, 60)
21	vase	(191, 66, 82)
22	lamp	(57, 129, 229)
23	toy	(158, 210, 31)
24	hinge	(190, 66, 191)
25	mirror	(57, 229, 178)
26	range hood	(210, 105, 31)
27	handle	(82, 66, 191)
28	appliance	(86, 229, 57)
29	jacket	(210, 31, 114)
30	air conditioner	(66, 160, 191)
31	robot	(229, 221, 57)
32	rod	(150, 31, 210)
33	pole	(66, 191, 113)
34	bag	(229, 70, 57)
35	laptop	(31, 69, 210)
36	closet	(129, 191, 66)
37	light switch	(229, 57, 194)
38	blanket	(31, 210, 194)
39	fabric	(191, 144, 66)
40	bed frame	(113, 57, 229)
41	bedding	(31, 210, 38)
42	fixture	(191, 66, 98)
43	light fixture	(57, 151, 229)
44	printer	(181, 210, 31)
45	cord	(174, 66, 191)
46	toilet	(57, 229, 156)
47	toilet seat	(210, 82, 31)
48	shoe	(66, 67, 191)
49	picture frame	(108, 229, 57)
50	bin	(210, 31, 137)
51	person	(66, 176, 191)
52	wheelchair	(229, 199, 57)
53	speaker	(127, 31, 210)
54	heater	(66, 191, 97)
55	microwave	(229, 57, 65)
56	dishwasher	(31, 92, 210)
57	3d printer	(145, 191, 66)
58	radiator	(229, 57, 216)
59	vacuum cleaner	(31, 210, 171)
60	tripod	(191, 128, 66)
61	drawer	(91, 57, 229)
62	cardboard	(47, 210, 31)
63	power strip	(191, 66, 114)
64	suitcase	(57, 173, 229)
65	coat	(204, 210, 31)
66	table	(159, 66, 191)
67	whiteboard	(57, 229, 134)
68	object	(210, 59, 31)
69	paper	(66, 83, 191)
70	backpack	(130, 229, 57)
71	coat rack	(210, 31, 159)
72	sign	(66, 191, 190)
73	monitor arm	(229, 177, 57)
74	mat	(104, 31, 210)
75	cushion	(66, 191, 81)
76	electronic	(229, 57, 87)
77	blind	(31, 115, 210)
78	basket	(161, 191, 66)
79	cloth	(220, 57, 229)
80	coffee maker	(31, 210, 148)
81	toilet paper	(191, 112, 66)
82	power outlet	(69, 57, 229)
83	post	(70, 210, 31)
84	container	(191, 66, 130)
85	drill	(57, 195, 229)
86	clothing	(210, 193, 31)
87	shelving unit	(143, 66, 191)
88	helmet	(57, 229, 112)
89	pillow	(210, 36, 31)
90	camera	(66, 99, 191)
91	equipment	(152, 229, 57)
92	gpu	(210, 31, 182)
93	water dispenser	(66, 191, 174)
94	computer	(229, 155, 57)
95	cable	(81, 31, 210)
96	pipe	(68, 191, 66)
97	book	(229, 57, 109)
98	shelf	(31, 138, 210)
99	rug	(177, 191, 66)
100	paper towel	(198, 57, 229)
101	dumbbell	(31, 210, 126)
102	computer tower	(191, 96, 66)
103	storage container	(57, 66, 229)
104	towel	(93, 210, 31)
105	decoration	(191, 66, 146)
106	bottle	(57, 217, 229)
107	toolbox	(210, 170, 31)
108	workbench	(127, 66, 191)
109	picture	(57, 229, 90)
110	shade	(210, 31, 49)
111	clamp	(66, 115, 191)
112	stand	(174, 229, 57)
113	adapter	(210, 31, 205)
114	device	(66, 191, 158)
115	floor	(229, 133, 57)
116	wall	(58, 31, 210)

Files

• annotations/train.json: COCO format annotations
• labelmap.txt: Class definitions with colors

Statistics

• Samples: 224
• Classes: 117
• Split: train