一种多骨架特征前期融合的在线行为识别算法
AN ONLINE SKELETON-BASED ACTION RECOGNITION ALGORITHM WITH MULTI-FEATURE EARLY FUSION
刘均发 1黎奕辉1
作者信息
- 1. 广东工业大学机电工程学院 广东广州 510006
- 折叠
摘要
针对现有基于人体骨架的行为识别方法存在计算量大、不适合在线应用的问题,提出一种多骨架特征前期融合的在线行为识别算法.该算法通过前期嵌入层融合不同类型的输入特征,并结合最大池化和层次池化操作提取骨架空间的多语义信息.根据日常行为的数据特征设计有效的骨架序列选取方式,并制作NTU-GAST Skeleton数据集,实现在线的行为识别应用.在公开数据集NTU60/120 RGB+D上进行测试,结果表明提出的算法需要更少计算量的同时取得了较高的识别准确率.
Abstract
Existing skeleton-based action recognition methods require large computation,which makes them unsuitable for online applications.Aiming at this problem,this paper proposes an online skeleton-based action recognition method with multi-feature early fusion.The algorithm integrated different types of input feature through the early embedding layer and combined the max pooling and hierarchical pooling to extract multi-semantic spatial information.The selection strategy of skeleton sequences was designed based on the characteristics of daily actions.A new 3D skeleton dataset,NTU-GAST Skeleton,was made for online action recognition.Experiments on NTU60 and 120 RGB+D dataset indicate that the proposed method achieves higher recognition accuracy with less computational complexity.
关键词
多骨架特征/前期融合/行为识别/在线应用Key words
Multi-skeleton feature/Early fusion/Action recognition/Online application引用本文复制引用
基金项目
广东省前沿与关键技术创新专项项目(2017B050506008)
广东省重点领域研发计划项目(2019B090915001)
出版年
2024
国家科技期刊平台
NSTL
维普
万方数据