电子科技大学学报2024，Vol.53Issue(1)：60-66,7.DOI:10.12178/1001-0548.2022401
面向骨架手势识别的全局时空可变形网络
Global Spatio-Temporal Deformable Network for Skeleton-Based Gesture Recognition
摘要
Abstract
The key of gesture recognition based on skeleton sequence is how to fuse spatio-temporal information and extract discriminate features.This paper proposes a key point focusing module.Through the global context modeling and the convolution method not limited to the fixed form,the network can span multiple frames and irrelevant key points,adaptively aggregate key point information closely related to gesture actions in the global scope,and extract the spatio-temporal characteristics of gesture.Experiments on Chalearn2013 and SHREC datasets show that the accuracy of our proposed method can reach 94.88% and 95.23%,and the method outperforms state-of-the-art methods.In addition,the method has better stability in dealing with noisy data and dynamic gestures.关键词
手势识别/特征提取/可变形卷积/骨架序列/全局信息Key words
gesture recognition/features extraction/deformable convolution/skeleton sequence/global information分类
信息技术与安全科学引用本文复制引用
石东子,林宏辉,刘一江,张鑫..面向骨架手势识别的全局时空可变形网络[J].电子科技大学学报,2024,53(1):60-66,7.基金项目
中央高校基本科研业务费交叉学科研究项目(2022ZYGXZR104) （2022ZYGXZR104）
广东省数字孪生人重点实验室项目(2022B1212010004) （2022B1212010004）
