Research on behavior recognition algorithm based on SE-I3D-GRU network

• 作者: Wu Jin Yang Xue Xi Meng Wan Xianghong
• 来源期刊: 高技术通讯（英文版）
• 年，卷（期）: 2021,（2）
• 页码范围: 163-172
• 页数: 10页
• 语种: 英文
• DOI: 10.3772/j.issn.1006-6748.2021.02.007

摘要

In order to effectively solve the problems of low accuracy and large amount of calculation of cur-rent human behavior recognition,a behavior recognition algorithm based on squeeze-and-excitation network(SENet)combined with 3D Inception network(I3D)and gated recurrent unit(GRU)net-work is proposed.The algorithm first expands the Inception module to three-dimensional,and builds a network based on the three-dimensional module,and expands SENet to three-dimensional,making it an attention mechanism that can pay attention to the three-dimensional channel.Then SENet is in-troduced into the I3D network,named SE-I3D,and SENet is introduced into the GRU network,named SE-GRU.And,SE-I3D and SE-GRU are merged,named SE-I3D-GRU.Finally,the net-work uses Softmax to classify the results in the UCF-101 dataset.The experimental results show that the SE-I3D-GRU network achieves a recognition rate of 93.2％on the UCF-101 dataset.