Vision-by-Language for Training-Free Compositional Image Retrieval.

Shyamgopal Karthik, Karsten Roth, Massimiliano Mancini, Zeynep Akata

Browse the full ICLR paper archive.