In recent years, human motion analysis has become the center of video processing, especially in motion detection and security surveillance. When performing specific tasks, it is always a tool to minimize human resources. However, earlier methods of detection often descriptors are extracted around spatiotemporal attention points or statistical features of motion regions are extracted, resulting in limited ability to effectively detect violence in videos. This paper is intended to use a Pre-trained Convolutional Neural Network Model (ResNet), where the video stream is sent to the ResNet together with a single shot detector (S S D) to detect violence and objects in the video. In addition, if an intense incident is identified, the relevant departments will be notified. This proposed methodology would be beneficial in monitoring scenarios such as detention center & asylum. Also, on mobile phones with cameras.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Violence Detection System Using Resnet


    Contributors:
    Shripriya, C (author) / Akshaya, J (author) / Sowmya, R (author) / Poonkodi, M (author)


    Publication date :

    2021-12-02


    Size :

    2243132 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    ResNet-RTDL-HPO_steps.tar.gz

    Holzmüller, David / Grinsztajn, Léo / Steinwart, Ingo | DataCite | 2024


    SIAMESE-RESNET: IMPLEMENTING LOOP CLOSURE DETECTION BASED ON SIAMESE NETWORK

    Qiu, Kai / Ai, Yunfeng / Tian, Bin et al. | British Library Conference Proceedings | 2018



    Siamese-ResNet: Implementing Loop Closure Detection based on Siamese Network

    Qiu, Kai / Ai, Yunfeng / Tian, Bin et al. | IEEE | 2018


    Motion Prediction for Autonomous Vehicles Using ResNet-Based Model

    ZeHao, Yao / Wang, LiQian / Liu, Ke et al. | IEEE | 2021