google / diffseg

DiffSeg is an unsupervised zero-shot segmentation method using attention information from a stable-diffusion model. This repo implements the main DiffSeg algorithm and additionally includes an experimental feature to add semantic labels to the masks based on a generated caption.
https://sites.google.com/corp/view/diffseg
MIT License
267 stars 21 forks source link

Question about the extension of Adding Semantics #8

Open Hugo-cell111 opened 6 months ago

Hugo-cell111 commented 6 months ago

Hi! In the appendix there is a chapter about Adding Semantics, but I haven't found any codes related to this chapter. Could you please upload the codes? Thanks!

PotatoTian commented 6 months ago

Hi, The code for adding semantics is included in the diffseg.ipynb file. You just need to run cell 1b to enable semantics.