TokenCut
TokenCut copied to clipboard
(CVPR 2022) Pytorch implementation of "Self-supervised transformers for unsupervised object discovery using normalized cut"
Hey there! Love the paper. Is it real-time capable of doing segmentation on mobile devices? Thanks!
When Using Vit-s for fine-tuning DINO on CUB, the descriptions between code and that in readme are not consistent. In code, about avgpool_patchtokens, typically set this to False for ViT-Small,...
Im trying to apply TokenCut to a dataset in which there is a combination of images without salient objects (flat, relatively textureless images), and images in which there are clear...
Thank you very much for your contribution. I don't quite understand the paragraph in your paper. Could you please explain it? The original text is as follows: Intuitively, di <...
I cannot find the "train" func. Am I supposed to write it on my own?