Open hohonuuli opened 1 year ago
I'd suggest renaming the issue to reflect our interest in using other foundation models, both purely visual ones like SAM and joint visual-language models. This will perhaps be best served by trying to store image embeddings for a given model.
@eor314 Maybe we should open a new issue? this one is very specific for just deploying SAM.
From @kakanikatija: