A simple and effective static gesture recognition method based on attention mechanism

被引:8
作者
Zhang, Lizao [1 ]
Tian, Qiuhong [1 ]
Ruan, Qionglu [1 ]
Shi, Zhixiang [1 ]
机构
[1] Zhejiang Sci Tech Univ, Sch Informat Sci & Technol, Hangzhou 310018, Peoples R China
基金
中国国家自然科学基金;
关键词
Sign language recognition; Deep learning; Attention mechanism;
D O I
10.1016/j.jvcir.2023.103783
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
To solve the problem of low sign language recognition rate under the condition of small samples, a simple and effective static gesture recognition method based on an attention mechanism is proposed. The method proposed in this paper can enhance the features of both the details and the subject of the gesture image. The input of the proposed method depends on the intermediate feature map generated by the original network. Also, the proposed convolutional model is a lightweight general module, which can be seamlessly integrated into any CNN(Convolutional Neural Network) architecture and achieve significant performance gains with minimal overhead. Experiments on two different datasets show that the proposed method is effective and can improve the accuracy of sign language recognition of the benchmark model, making its performance better than the existing methods.
引用
收藏
页数:9
相关论文
共 50 条
[41]   An attention-based RGBD dual-branch gesture recognition network [J].
Chen, Bo ;
Xie, Pengwei ;
Hao, Nan .
2021 PROCEEDINGS OF THE 40TH CHINESE CONTROL CONFERENCE (CCC), 2021, :8022-8027
[42]   Human Action Recognition Method Based on Video-Level Features and Attention Mechanism [J].
Cai, Qiang ;
Yan, Jin ;
Li, Haisheng ;
Deng, Yibiao .
PROCEEDINGS OF 2020 CHINESE INTELLIGENT SYSTEMS CONFERENCE, VOL I, 2021, 705 :225-233
[43]   Research on named entity recognition method of marine natural products based on attention mechanism [J].
Ma, Xiaodong ;
Yu, Rilei ;
Gao, Chunxiao ;
Wei, Zhiqiang ;
Xia, Yimin ;
Wang, Xiaowei ;
Liu, Hao .
FRONTIERS IN CHEMISTRY, 2023, 11
[44]   Lightweight facial expression recognition method based on attention mechanism and key region fusion [J].
Kong, Yinghui ;
Ren, Zhaohan ;
Zhang, Ke ;
Zhang, Shuaitong ;
Ni, Qiang ;
Han, Jungong .
JOURNAL OF ELECTRONIC IMAGING, 2021, 30 (06)
[45]   Gait Recognition in Different Terrains with IMUs Based on Attention Mechanism Feature Fusion Method [J].
Mengxue Yan ;
Ming Guo ;
Jianqiang Sun ;
Jianlong Qiu ;
Xiangyong Chen .
Neural Processing Letters, 2023, 55 :10215-10234
[46]   Gait Recognition in Different Terrains with IMUs Based on Attention Mechanism Feature Fusion Method [J].
Yan, Mengxue ;
Guo, Ming ;
Sun, Jianqiang ;
Qiu, Jianlong ;
Chen, Xiangyong .
NEURAL PROCESSING LETTERS, 2023, 55 (08) :10215-10234
[47]   Human body gesture recognition method based on deep learning [J].
Chen, Liang ;
Li, Yue ;
Liu, Yunting .
PROCEEDINGS OF THE 32ND 2020 CHINESE CONTROL AND DECISION CONFERENCE (CCDC 2020), 2020, :587-591
[48]   An effective deep learning method with multi-feature and attention mechanism for recognition of Chinese rice variety information [J].
Helong Yu ;
Ziqing Li ;
Chunguang Bi ;
Huiling Chen .
Multimedia Tools and Applications, 2022, 81 :15725-15745
[49]   An effective deep learning method with multi-feature and attention mechanism for recognition of Chinese rice variety information [J].
Yu, Helong ;
Li, Ziqing ;
Bi, Chunguang ;
Chen, Huiling .
MULTIMEDIA TOOLS AND APPLICATIONS, 2022, 81 (11) :15725-15745
[50]   Abnormal Behavior Recognition Based on Hybrid Attention Mechanism [J].
Sun, Xiaohu ;
Yu, Axiang ;
Shen, Xulin ;
Li, Hongjun .
Computer Engineering and Applications, 2024, 59 (05) :140-147