search

wusize / CLIPSelf

[ICLR2024 Spotlight] Code Release of CLIPSelf: Vision Transformer Distills Itself for Open-Vocabulary Dense Prediction
https://arxiv.org/abs/2310.01403
Other
161 stars 9 forks source link

Can you provide the VIT/B 16 weights trained by CLIPself on the COCO dataset using Open AI clip? #2

Closed winnerwu6 closed 11 months ago

winnerwu6 commented 11 months ago

I want to use this weight, but my GPU is not enough for me to train it, I would be grateful if you could provide it！！！

wusize commented 11 months ago

Hi! Please find the weights under the link. But personally I suggest the use of EVA-CLIP models for higher performance and efficiency.

winnerwu6 commented 11 months ago

Thank you for your reply, I appreciate it very much！