[CVPR-2022] Official implementations of CIRKD: Cross-Image Relational Knowledge Distillation for Semantic Segmentation and implementations on Cityscapes, ADE20K, COCO-Stuff., Pascal VOC and CamVid.
I am writing to you with one question.
If you look at github, Test mIoU for the Cityscapes dataset is listed, but there is no list of gtFine in test.lst, so I wonder how you experimented.
Thanks for the good research.
I am writing to you with one question. If you look at github, Test mIoU for the Cityscapes dataset is listed, but there is no list of gtFine in test.lst, so I wonder how you experimented.