1 paper
Minghui Lin, Shu Wang, Xiang Wang +4
Current multi-modal object re-identification approaches based on large-scale pre-trained backbones (i.e., ViT) have displayed remarkable progress and achieved excellent performance…