通过对比视图不变表示进行跨视图动作识别（英文版）

英文标题：Cross-view Action Recognition via Contrastive View-invariant Representation中文摘要：提出了一种简单高效的跨视角动作识别（CVAR）框架，能够从 RGB 视频、3D 骨架数据或两者学习不变特征。在 N-UCLA、NTU-RGB+D 60、NTU-RGB+D 120 和 UWA3DII 数据集上，该方法表现出优异的性能，分别达到了 99.4％（RGB）、99.9％（3D 骨架）英文摘要：Cross view action recognition (CVAR) seeks to recognize a human action whenobserved from a previously unseen viewpoint. This is a challenging problemsince the appearance of an action changes significantly with the viewpoint.Applications of CVAR include surveillan