CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
Nobody has forked this repository yet.