<- Back to Datasets

PASS

An ImageNet replacement for self-supervised pretraining without humans

PASS

PASS is a large-scale image dataset that does not include any humans and which can be used for high-quality pretraining while significantly reducing privacy concerns.

View this Dataset
->
Visual Geometry Group, University of Oxford
View author website
Task
Image Classification
Annotation Types
Classification Tags
6000000
Items
20000
Classes
6000000
Labels
Models using this dataset
Last updated on 
January 20, 2022
Licensed under 
CC-BY
Gain control of your training data
15,000+ ML engineers can’t be wrong