skip to main content
10.1145/3448734.3450778acmotherconferencesArticle/Chapter ViewAbstractPublication PagesdtmisConference Proceedingsconference-collections
research-article

Research on Human Behavior Recognition Based on Video Key Frame

Published: 17 May 2021 Publication History

Abstract

In order to solve the problems of low recognition accuracy and high computational complexity caused by redundant video data in the existing behavior recognition process, a human behavior recognition method based on video key frame (S3DCCA) is proposed. First of all, structural similarity (SSIM) algorithm is used to calculate the difference of luminance, contrast and structure between the two frames, and the result is multiplied to attain SSIM value, then select the local and global key frame in the human motion video frame sequence according to the SSIM value. Finally, the selected key frame are used as the input of three-dimensional convolutional neural networks and attention mechanism Channel attention (3DCCA) model to recognize human behavior. Experimental results on UCF101 and HMDB51 datasets show that the proposed method has high recognition rate.

References

[1]
Kim J H, Cho Y I. A New Residual Attention Network based on Attention Models for Human Action Recognition in Video[J]. Journal of the Korea Society of Computer and Information, 2020, 25(1): 55-61.
[2]
Ye M Q, Research on Theory and Method of Human Behavior Recognition Based on Video[D]. Chengdou: University of Electronic Science and Technology of China, 2018.
[3]
He D. Multi-feature fusion behavior recognition system based on key frame[D]. Xian: Northwestern Polytechnical University, 2019.
[4]
Zhang X D, Yang J, Hu L M, Human motion recognition with time-of-flight camera based on multi-layer motion history images[J]. Journal of electronics and information, 2014, 36(5): 1139-1144.
[5]
Song J M, Zhang Y, Gao Z, Human motion description algorithm based on deep and dense space-time points of interest. Pattern recognition and artificial intelligence, 2015, 28(10): 939-945.
[6]
Vemulapalli R, Arrate F, Chellappa R. Human Action Recognition by Representing 3D Human Skeletons as Points in a Lie Group[C]//proceedings of the IEEE Conference on Computer Vision & Pattern Recognition. 2014: 588-595.
[7]
Du Y, Wang W, Wang L. Hierarchical recurrent neural network for skeleton based action recognition[C]//proceedings of the Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 1110-1118.
[8]
Wang H, Schmid C. Action recognition with improved trajectories[C]//Proceedings of the IEEE international conference on computer vision. 2013: 3551-3558.
[9]
Simonyan K, Zisserman A. Two-stream convolutional networks for action recognition in videos[C]//Advances in neural information processing systems. 2014: 568-576.
[10]
Yue-Hei Ng J, Hausknecht M, Vijayanarasimhan S, Beyond short snippets: Deep networks for video classification[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 4694-4702.
[11]
Chen C, Jafari R, Kehtarnavaz N. Action Recognition from Depth Sequences Using Depth Motion Maps-Based Local Binary Patterns[C]//proceedings of the 2015 IEEE Winter Conference on Applications of Computer Vision. 2015: 1092-1099.
[12]
Salih AAA, Youssef C. Spatiotemporal representation of 3D skeleton joints-based action recognition using modified spherical harmonics[J]. Pattern Recognition Letters, 2016, 83: 32-41.
[13]
Tran D, Bourdev L, Fergus R, Learning spatiotemporal features with 3d convolutional networks[C]//Proceedings of the IEEE international conference on computer vision. 2015: 4489-4497.
[14]
Zhou W, Bovik A C, Sheikh H R, Image Quality Assessment: From Error Visibility to Structural Similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4): 660-612.

Index Terms

  1. Research on Human Behavior Recognition Based on Video Key Frame
        Index terms have been assigned to the content through auto-classification.

        Recommendations

        Comments

        Information & Contributors

        Information

        Published In

        cover image ACM Other conferences
        CONF-CDS 2021: The 2nd International Conference on Computing and Data Science
        January 2021
        1142 pages
        ISBN:9781450389570
        DOI:10.1145/3448734
        Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

        Publisher

        Association for Computing Machinery

        New York, NY, United States

        Publication History

        Published: 17 May 2021

        Permissions

        Request permissions for this article.

        Check for updates

        Author Tags

        1. 3DCNN
        2. Deep learning
        3. attention mechanism
        4. behavior recognition
        5. key frame

        Qualifiers

        • Research-article
        • Research
        • Refereed limited

        Conference

        CONF-CDS 2021

        Contributors

        Other Metrics

        Bibliometrics & Citations

        Bibliometrics

        Article Metrics

        • 0
          Total Citations
        • 38
          Total Downloads
        • Downloads (Last 12 months)3
        • Downloads (Last 6 weeks)2
        Reflects downloads up to 20 Feb 2025

        Other Metrics

        Citations

        View Options

        Login options

        View options

        PDF

        View or Download as a PDF file.

        PDF

        eReader

        View online with eReader.

        eReader

        HTML Format

        View this article in HTML Format.

        HTML Format

        Figures

        Tables

        Media

        Share

        Share

        Share this Publication link

        Share on social media