A collection of open and reproducible self-supervised Vision Transformer (ViT) checkpoints trained exclusively on ImageNet-1K (1.43M images).
-
SSL ViT Patch PCA Visualiser
👀1Visualize ViT patch features with PCA for any image
-
OK-AI/ibot-vits16-pretrain-in1k
Image Feature Extraction • 44M • Updated • 66 -
OK-AI/lejepa-vits16-pretrain-in1k
Image Feature Extraction • 26.8M • Updated • 48 -
OK-AI/dino-vits16-pretrain-in1k
Image Feature Extraction • 44M • Updated • 374