Dataset

COCO-164k

Name: COCO-164k
Creator: Pictograph Research
Published: 2026-07-03T14:27:57.085619+00:00
License: https://creativecommons.org/licenses/by/4.0/

Pictograph Research · Updated 2mo ago · CC-BY-4.0

1 Sign in to import

Details

163,957Images

80Classes

1Stars

Annotations885,976
Typespolygon · bbox
Updated2mo ago
OrganizationPictograph Research
LicenseCC-BY-4.0

Classes 80

person 267,956 car 45,410 chair 39,844 book 25,204 bottle 25,082 cup 21,469 dining table 16,390 bowl 14,946 traffic light 13,429 handbag 12,875 umbrella 11,672 boat 10,978 bird 10,966 truck 10,367 bench 10,218 sheep 9,571 banana 9,565 kite 9,129 backpack 9,085 motorcycle 9,020 potted plant 8,971 cow 8,386 wine glass 8,180 carrot 8,123 knife 8,085 broccoli 7,573 bicycle 7,367 donut 7,333 skis 6,864 vase 6,851 horse 6,839 tie 6,700 cell phone 6,681 cake 6,606 clock 6,587 orange 6,587 sports ball 6,559 spoon 6,412 suitcase 6,411 surfboard 6,362 bus 6,317 pizza 6,091 tv 6,091 couch 6,040 apple 6,012 remote 5,983 sink 5,834 elephant 5,736 dog 5,717 skateboard 5,715 fork 5,689 zebra 5,535 giraffe 5,333 airplane 5,264 laptop 5,191 tennis racket 5,032 cat 4,967 teddy bear 4,919 train 4,760 sandwich 4,533 bed 4,355 toilet 4,328 baseball glove 3,895 oven 3,477 baseball bat 3,418 hot dog 3,009 keyboard 3,007 frisbee 2,796 refrigerator 2,760 snowboard 2,750 mouse 2,367 stop sign 2,058 toothbrush 2,002 fire hydrant 1,964 microwave 1,727 scissors 1,500 bear 1,365 parking meter 1,343 toaster 234 hair drier 209

Dataset Card

COCO-164k

A public computer-vision dataset of 163,957 images with 885,976 annotations across 80 classes, annotated as polygon and bounding box.

Pictograph Research · CC-BY-4.0 · Public · Updated 2026-08-08

Overview

This dataset holds 163,957 images annotated as polygon and bounding box across 80 classes. The most common auto-detected tags are playing, eating, person, eye-level view, top-down view, zoo. It is ready to fork, export in a dozen formats, or train a model from.

At a glance

Metric	Value
Images	163,957
Annotations	885,976
Classes	80
Annotation types	polygon, bounding box
Folders	3
Storage	Standard
Visibility	Public

What's in these images

Tags auto-detected across the dataset, most common first.

Tag	Category	Images
playing	Attributes	22,448
eating	Attributes	17,689
person	Objects	16,034
eye-level view	Attributes	12,679
top-down view	Attributes	10,642
zoo	Scenes	8,089
walking	Attributes	7,380
driving	Attributes	7,018
child	Objects	6,740
sign	Objects	6,474
cooking	Attributes	6,465
bathroom	Scenes	6,319
ski	Objects	6,274
skiing	Attributes	6,248
jumping	Attributes	6,024

Use this dataset

from pictograph import Client, Dataset

client = Client()  # reads PICTOGRAPH_API_KEY
dataset: Dataset = client.datasets.get("COCO-164k")
print(dataset.image_count, dataset.annotation_count)