Multi-View Action Recognition using Contrastive Learning

WACV(2023)

引用 3|浏览29
暂无评分
摘要
In this work, we present a method for RGB-based action recognition using multi-view videos. We present a supervised contrastive learning framework to learn a feature embedding robust to changes in viewpoint, by effectively leveraging multi-view data. We use an improved supervised contrastive loss and augment the positives with those coming from synchronized viewpoints. We also propose a new approach to use classifier probabilities to guide the selection of hard negatives in the contrastive loss, to learn a more discriminative representation. Negative samples from confusing classes based on posterior are weighted higher. We also show that our method leads to better domain generalization compared to the standard supervised training based on synthetic multi-view data. Extensive experiments on real (NTU-60, NTU-120, NUMA) and synthetic (RoCoG) data demonstrate the effectiveness of our approach.
更多
查看译文
关键词
recognition,action,learning,multi-view
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要