A Study on Video Anomalous Behavior Recognition Based on Two-Stream Network
Xiaodong Luo, Ying Liu, Yu Hao, Huimin Du · 2024
With the wide application of surveillance systems and the continuous emergence of large-scale video data, video anomalous behavior recognition has become particularly important in the fields of security monitoring, intelligent transportation, industrial production, etc.Two-stream CNN combines optical flow streaming and frame difference streaming to better capture the spatio-temporal information in the video and improve the model performance. In order to further improve the accuracy and robustness of video anomalous behavior recognition, we introduced the CBAM (Convolutional Block Attention Module) module in this study. By fusing CBAM into a two-stream CNN, we aim to enhance the model's ability to perceive anomalous behaviors in videos and improve its recognition performance.