Can we fine-tune/retrain this model for specific image processing domains such as segmenting dense cell cluster in biomedical imaging？

[lgn211]

As titled.

[Xushuolin]

Absolutely

[dstoekl]

see https://github.com/facebookresearch/segment-anything/issues/5

[JunMa11]

@lgn211 @Xushuolin

We provide a step-by-step tutorial on fine-tuning SAM on 2D and 3D medical image datasets. It requires less than 10G GPU memory. Hope that it could be useful.

https://github.com/bowang-lab/MedSAM#model-training-video-tutorial