Code for our paper : Mixed-supervised segmentation: Confidence maximization helps knowledge distillation. https://arxiv.org/abs/2109.10902
Primary LanguagePythonMIT LicenseMIT