LiheYoung / UniMatch

[CVPR 2023] Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation
https://arxiv.org/abs/2208.09910
MIT License
456 stars 59 forks source link

Can UniMatch be used for Domain generalization #63

Closed wangmingaaaaa closed 1 year ago

wangmingaaaaa commented 1 year ago

Great work! Is it effective for generalizing medical image segmentation in the semi-supervised domain? Any suggestions? would it work?

LiheYoung commented 1 year ago

I have not tried this. But I think in the semi-supervised DG, if you previously used a single strongly augmented image, you can try to use two strong streams just as UniMatch does. It probably helps.