KS-FuseNet: An Efficient Action Recognition Method Based on Keyframe Selection and Feature Fusion

被引:0
|
作者
Mao, Keming [1 ]
Xiao, Yilong [1 ]
Jing, Xin [1 ]
Hu, Zepeng [1 ]
Ping, Yi [1 ]
机构
[1] Northeastern Univ, Software Coll, Shenyang, Peoples R China
来源
PATTERN RECOGNITION AND COMPUTER VISION, PRCV 2024, PT VII | 2025年 / 15037卷
关键词
Action recognition; Spatial-temporal; Feature fusion; Keyframe selection; CONTEXT;
D O I
10.1007/978-981-97-8511-7_38
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Addressing the challenge of effectively capturing features in contemporary video tasks, we propose an action recognition approach grounded in keyframe filtering and feature fusion. Our method comprises two core modules. The keyframe screening module employs an attention mechanism to segregate the input depth feature map sequence into two distinct tensors, effectively reducing spatial redundancy computation and enhancing key feature capture. The other spatio-temporal and action feature module features two branches with divergent structures, performing spatio-temporal and action feature extraction on the differentiated features from the previous module. Through these closely linked modules, our approach effectively discerns and extracts meaningful video features for subsequent classification tasks. We construct an end-to-end deep learning model using established frameworks, training and validating it on a generic video dataset, and confirm its efficacy through comparison and ablation experiments. Experiments conducted on this dataset demonstrate that our model surpasses the majority of prior works.
引用
收藏
页码:540 / 553
页数:14
相关论文
共 50 条
  • [31] Fusion of feature selection methods in gene recognition
    Gil, Fabian
    Osowski, Stanislaw
    BULLETIN OF THE POLISH ACADEMY OF SCIENCES-TECHNICAL SCIENCES, 2021, 69 (03)
  • [32] TRAJECTORY FEATURE FUSION FOR HUMAN ACTION RECOGNITION
    Megrhi, Sameh
    Beghdadi, Azeddine
    Souidene, Wided
    2014 5TH EUROPEAN WORKSHOP ON VISUAL INFORMATION PROCESSING (EUVIP 2014), 2014,
  • [33] Feature and Decision Level Fusion for Action Recognition
    Abouelenien, Mohamed
    Wan, Yiwen
    Saudagar, Abdullah
    2012 THIRD INTERNATIONAL CONFERENCE ON COMPUTING COMMUNICATION & NETWORKING TECHNOLOGIES (ICCCNT), 2012,
  • [34] An efficient bit-based feature selection method
    Chen, Wei-Chou
    Tseng, Shian-Shyong
    Hong, Tzung-Pei
    EXPERT SYSTEMS WITH APPLICATIONS, 2008, 34 (04) : 2858 - 2869
  • [35] Bearing Fault Feature Selection Method Based on Weighted Multidimensional Feature Fusion
    Li, Yazhou
    Dai, Wei
    Zhang, Weifang
    IEEE ACCESS, 2020, 8 : 19008 - 19025
  • [36] Face recognition method based on sparse representation and feature fusion
    Jiang, Changjiang
    Wang, Mingyi
    Tang, Xianlun
    Mao, Rong
    2019 CHINESE AUTOMATION CONGRESS (CAC2019), 2019, : 396 - 400
  • [37] Face recognition based on a new nonlinear feature fusion method
    Zhao, Feng
    Yang, Yinling
    Ma, Ruichuan
    Yuan, Da
    Journal of Information and Computational Science, 2015, 12 (07): : 2613 - 2621
  • [38] Recognition Method of Orchard Unstructured Road Based on Feature Fusion
    Zhang Y.
    Feng Z.
    Zhang J.
    Gong J.
    Lan Y.
    Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery, 2023, 54 (07): : 35 - 44+67
  • [39] An Identity Recognition Method Based on ElectroCardioGraph and PhotoPlethysmoGraph Feature Fusion
    Xiao Jian
    Li Sizhuo
    Dong Wei
    Li Qinghua
    Hu Fang
    JOURNAL OF ELECTRONICS & INFORMATION TECHNOLOGY, 2021, 43 (10) : 3010 - 3017
  • [40] Music emotion recognition method based on multi feature fusion
    Zhang, Yali
    INTERNATIONAL JOURNAL OF ARTS AND TECHNOLOGY, 2022, 14 (01) : 10 - 23