Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (8): 2551-2557.DOI: 10.11772/j.issn.1001-9081.2023081157

• Multimedia computing and computer simulation • Previous Articles     Next Articles

Video pedestrian anomaly detection method based on skeleton graph and mixed attention

Yuhan LIU, Genlin JI(), Hongping ZHANG   

  1. School of Computer and Electronic Information / School of Artificial Intelligence,Nanjing Normal University,Nanjing Jiangsu 210023,China
  • Received:2023-08-29 Revised:2023-09-20 Accepted:2023-10-08 Online:2024-08-22 Published:2024-08-10
  • Contact: Genlin JI
  • About author:LIU Yuhan, born in 1999, M. S. candidate. Her research interests include big data analysis and mining, video anomaly detection.
    ZHANG Hongping, born in 1999, M. S. candidate. Her research interests include pattern recognition, machine learning.
  • Supported by:
    National Natural Science Foundation of China(41971343)

基于骨架图与混合注意力的视频行人异常检测方法

刘禹含, 吉根林(), 张红苹   

  1. 南京师范大学 计算机与电子信息学院/人工智能学院,南京 210023
  • 通讯作者: 吉根林
  • 作者简介:刘禹含(1999—),女(满族),吉林吉林人,硕士研究生,CCF会员,主要研究方向:大数据分析与挖掘、视频异常检测
    吉根林(1964—),男,江苏海安人,教授,博士生导师,博士,CCF会员,主要研究方向:大数据分析与挖掘 glji@njnu.edu.cn
    张红苹(1999—),女,浙江湖州人,硕士研究生,主要研究方向:模式识别、机器学习。
  • 基金资助:
    国家自然科学基金资助项目(41971343)

Abstract:

In recent years, many studies that use human skeleton graph for video anomaly detection only consider the directly connected nodes when describing the strength of human skeleton connection, focusing on a small moving region and ignoring local features, so it is still very difficult to accurately detect pedestrian abnormal events. To solve the above problems, a video pedestrian anomaly detection method called PAD-SGMA (video Pedestrian Anomaly Detection method based on Skeleton Graph and Mixed Attention) was proposed. The association between skeleton points was extended, the root node was connected with the nodes that were not directly connected, and the human skeleton graph was divided to obtain the local features of the human skeleton. In the graph convolution module, static global skeleton, local region skeleton and attention-based adjacency matrix were used to capture the hierarchical representation. Secondly, a new convolutional network of spatio-temporal channel mixed attention was proposed, in which a mixed attention module was added to focus on spatial and channel relationships, to help the model enhance distinguishing features and pay different degrees of attention to different joints. In order to verify the proposed model, experiments were carried out on a large-scale open standard dataset ShanghaiTech Campus dataset, and the experimental results showed that the AUC(Area Under Curve) of PAD-SGMA was increased by 0.018 compared with GEPC (Graph Embedded Pose Clustering).

Key words: video anomaly detection, deep learning, human skeleton, graph convolutional network, attention

摘要:

近些年,许多利用人体骨架图检测视频异常的研究在描述人体骨架连接强弱时,只考虑到直接相连的节点,关注的运动区域较小且忽略了局部特征,很难准确检测行人异常事件。为解决以上问题,提出一种基于骨架图与混合注意力的视频行人异常检测方法(PAD-SGMA)。首先,扩展骨架点之间的关联,连接根节点与未直接相连的节点,并划分人体骨架图,获取人体骨架局部特征,在图卷积模块中利用静态全局骨架、局部区域骨架和基于注意的邻接矩阵来捕获层次表示;其次,提出新的时空通道混合注意图卷积网络,增加混合注意力模块,关注空间和通道关系,帮助模型增强区分特征且不同程度地关注每个关节。为了验证所提模型,在大规模的公开标准数据集ShanghaiTech Campus上进行实验,结果表明,与GEPC(Graph Embedded Pose Clustering)相比,PAD-SGMA的AUC(Area Under Curve)提高了0.018。

关键词: 视频异常检测, 深度学习, 人体骨架, 图卷积网络, 注意力

CLC Number: