Fetching the paper…

DFormer: Rethinking RGBD Representation Learning for Semantic Segmentation · Around