How to make LLM output object-centered visual annotations without decoding the LLM's output, so that it does not generate text but only produces segmented images.

lxtGH / OMG-Seg

OMG-LLaVA and OMG-Seg codebase [CVPR-24 and NeurIPS-24]

Other

1.32k stars 50 forks source link

How to make LLM output object-centered visual annotations without decoding the LLM's output, so that it does not generate text but only produces segmented images. #44

Open aimll101 opened 2 months ago

aimll101 commented 2 months ago

How to make LLM output object-centered visual annotations without decoding the LLM's output, so that it does not generate text but only produces segmented images.