Multiple instance-based video anomaly detection using deep temporal encoding-decoding

被引:26
|
作者
Kamoona, Ammar Mansoor [1 ,2 ]
Gostar, Amirali Khodadadian [1 ]
Bab-Hadiashar, Alireza [1 ]
Hoseinnezhad, Reza [1 ]
机构
[1] RMIT Univ, Sch Engn, Melbourne, Vic 3000, Australia
[2] Univ Kufa, Fac Engn, Dept Elect Engn, Najaf, Iraq
基金
澳大利亚研究理事会;
关键词
Anomaly detection; Surveillance videos; Weakly supervised multiple instance learning; EVENT DETECTION; BEHAVIOR DETECTION; CLASSIFICATION; LOCALIZATION; FRAMEWORK; NETWORKS;
D O I
10.1016/j.eswa.2022.119079
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper, we propose a weakly supervised deep temporal encoding-decoding solution for anomaly detection in surveillance videos using multiple instance learning. The proposed approach uses both abnormal and normal video clips during the training phase which is developed in the multiple instance learning framework where we treat the video as a bag and video clips as instances in the bag. Our main contribution lies in the proposed novel approach to consider temporal relations between video instances in a weakly supervised setting. We deal with video instances (clips) as sequential visual data rather than a set of independent instances. We employ a deep temporal encoding-decoding network that is designed to capture spatio-temporal evolution of video instances over time. We also propose a new loss function that maximizes the mean distance between normal and abnormal instance predictions. The new loss function ensures a low false alarm rate which is crucial in practical surveillance applications. The proposed temporal encoding-decoding approach with the modified loss is benchmarked against the state of the art. The results show that the proposed method performs similar to, or better than the state-of-the-art solutions for anomaly detection in video surveillance applications and achieves the lowest false alarm rate on UCF-Crime dataset.
引用
收藏
页数:13
相关论文
共 50 条
  • [21] A Robust Instance-based Video Search Method By Re-ranking for Multiple Visual Codebooks
    Wang, Wen
    Zhao, Zhicheng
    PROCEEDINGS OF THE 2012 EIGHTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND SECURITY (CIS 2012), 2012, : 163 - 167
  • [22] Video Instance Lane Detection via Deep Temporal and Geometry Consistency Constraints
    Wang, Mingqian
    Zhang, Yujun
    Feng, Wei
    Zhu, Lei
    Wang, Song
    MM 2022 - Proceedings of the 30th ACM International Conference on Multimedia, 2022, : 2324 - 2332
  • [23] Revolutionizing sentiment classification: A deep learning approach using self-attention based encoding-decoding transformers with feature fusion
    Tejashwini, S. G.
    Aradhana, D.
    ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2023, 125
  • [24] Video Instance Lane Detection via Deep Temporal and Geometry Consistency Constraints
    Wang, Mingqian
    Zhang, Yujun
    Feng, Wei
    Zhu, Lei
    Wang, Song
    PROCEEDINGS OF THE 30TH ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2022, 2022, : 2324 - 2332
  • [25] Multiple Instance Learning with Deep Instance Selection for Video-based Face Recognition
    Liu, Ning
    PROCEEDINGS OF THE 2016 3RD INTERNATIONAL CONFERENCE ON MECHATRONICS AND INFORMATION TECHNOLOGY (ICMIT), 2016, 49 : 327 - 332
  • [26] Prompt-Enhanced Multiple Instance Learning for Weakly Supervised Video Anomaly Detection
    Chen, Junxi
    Li, Liang
    Su, Li
    Zha, Zheng-Jun
    Huang, Qingming
    2024 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2024, : 18319 - 18329
  • [27] Video Anomaly Detection based on Deep Generative Network
    Saypadith, Savath
    Onoye, Takao
    2021 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS), 2021,
  • [28] Dy-MIL: dynamic multiple-instance learning framework for video anomaly detection
    Chen Li
    Mo Chen
    Multimedia Systems, 2024, 30
  • [29] Dy-MIL: dynamic multiple-instance learning framework for video anomaly detection
    Li, Chen
    Chen, Mo
    MULTIMEDIA SYSTEMS, 2024, 30 (01)
  • [30] A Self-Paced Multiple Instance Learning Framework for Weakly Supervised Video Anomaly Detection
    He, Ping
    Li, Huibin
    Han, Miaolin
    APPLIED SCIENCES-BASEL, 2025, 15 (03):