WHEN SKELETON MEETS APPEARANCE: ADAPTIVE APPEARANCE INFORMATION ENHANCEMENT FOR SKELETON BASED ACTION RECOGNITION
文献类型：会议论文
| 作者 | Wang Suqin2,3; Zhou Lu2,3 ; Chen Yingying1,2,3; Jiangtao Huo4; Wang Jinqiao2,3
|
| 出版日期 | 2022 |
| 会议日期 | July 18-22, 2022 |
| 会议地点 | 中国台北 |
| 英文摘要 | Skeleton-based action recognition methods which utilize
graph convolution networks (GCNs) have achieved remark
able success in recent years. However, action recognizer can
be easily confused by the ambiguity caused by different ac
tions with similar skeleton sequences when only skeleton data
is trained. Introducing appearance information can effectively
eliminate the ambiguity. Based on this, we introduce a two
stream network for action recognition. One trained on RGB
images extracts appearance information. The other trained
on skeleton data models motion information and adaptively
captures appearance information of action areas at action
related intervals via a specially tailored attention mechanism.
Our architecture is trained and evaluated on two large-scale
datasets: NTU RGB+D and NTU RGB+D 120, and a small
scale human-object interaction dataset Northwestern-UCLA.
Experiment results verify the effectiveness of our method and
the performance of our method exceeds the state-of-the-art
with a significant margin. |
| 源URL | [http://ir.ia.ac.cn/handle/173211/57151]  |
| 专题 | 紫东太初大模型研究中心 |
| 作者单位 | 1.Development Research Institute of Guangzhou Smart City 2.School of Artificial Intelligence, University of Chinese Academy of Sciences 3.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 4.Army Medical University, NCO School of PLA |
| 推荐引用方式 GB/T 7714 | Wang Suqin,Zhou Lu,Chen Yingying,et al. WHEN SKELETON MEETS APPEARANCE: ADAPTIVE APPEARANCE INFORMATION ENHANCEMENT FOR SKELETON BASED ACTION RECOGNITION[C]. 见:. 中国台北. July 18-22, 2022. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。