End-to-End Video Semantic Segmentation in Adverse Weather using Fusion Blocks and Temporal-Spatial Teacher-Student Learning Xin Y ang 1 Y an Wending 2 Michael Bi Mi2 Yuan

Oct-10-2025, 22:38:36 GMT–Neural Information Processing Systems 

The key idea of our fusion block is to offer the model a way to merge information from consecutive frames by matching and merging relevant pixels from those frames.