Gait recognition is a new technology that can identify people with various walking patterns. Gait of a person is the manner in which they walk. A person’s walking stride is distinctive due to their body movements. Due to the popularity of the Kinect, human gait can be recognised to using 3D skeletal information. In this paper, Skeleton Gait Energy(SkeGEI) is image-based approach for gait recognition and VGG-16 with multilayer perceptron is proposed to effectively exploit raw depth information collected by the Kinect sensor. To restore as much gait data as possible, fine-tuned VGG-16 is used to extract space - time deep feature data from SkeGEI. Multilayer perceptron is then used to ascertain the connection between the corresponding subject and the gait features. Softmax is used for classification. Experiments on three different datasets show that our technique outclasses the majority of gait recognition methods.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    VGG-16 based Gait Recognition using Skeleton Features


    Contributors:
    Ali, Irshad (author) / Junaid, Iman (author) / Ari, Samit (author)


    Publication date :

    2023-03-17


    Size :

    1043708 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Gait characterization using dynamic skeleton acquisition

    E. Gianaria / N. Balossino / M. Grangetto et al. | BASE | 2013

    Free access

    Robust gait identification using Kinect dynamic skeleton data

    Gianaria, Elena / Grangetto, Marco | BASE | 2019

    Free access

    Dynamic Distance-Based Shape Features for Gait Recognition

    Whytock, T. | British Library Online Contents | 2014


    Gait Analysis Method Focused on Skeleton Pose Estimation Using Wheeled Gait-Training Walker

    Nakayama, Hidehisa / Ogawa, Kazutoshi / Maruyama, Tsugito | TIBKAT | 2022


    Gait Analysis based on Azure Kinect 3D Human Skeleton

    Zeng, Yan / Wu, Liufan / Xie, Donglin | IEEE | 2021