Salient Attention Model and Classes Imbalance Remission for Video Anomaly Analysis with Weak Label

Zhou, Hang; Xia, Huifen; Zhan, Yongzhao; Mao, Qirong

doi:10.1007/978-3-030-70626-5_13

Cited by 2 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Song et al [3] utilized a multi-scale temporal feature fusion module to capture the temporal correlations of segments over different time spans. Research focusing on spatiotemporal correlations between video segments [4] suggested that considering global spatiotemporal correlation information could lead to the dilution of abnormal segment features by the large amount of normal segment feature information. This dilution could cause the model to either fail to accurately detect abnormal regions or to misclassify abnormal events as normal ones.…”

Section: Introductionmentioning

confidence: 99%

Normal-Abnormal Negative Impacts Suppressing Via Normal Feature Memory For Video Anomaly Detection

Chen,

Liang,

Zhan

2024

Preprint

View full text Add to dashboard Cite

In the field of video anomaly detection, effective representation for the normal and abnormal video features is a key issue for improving the accuracy of detection models. The feature representation based on global spatio-temporal networks lacks to consider negative impacts for the features between normal segment and abnormal segment, resulting in incorrect detection for some segments in video. To address this issue, a new approach of normal-abnormal negative impacts suppressing via normal feature memory for video anomaly detection is proposed. In this method, feature representation with normal–abnormal negative impacts suppression via normal feature memory pool is modeled, which can be used to suppress normal-abnormal negative impacts in video feature learning. Constraint learning for normal feature memory pool is designed to promote more effective learning for the normal feature memory pool and feature of abnormal video segment. Extensive experiments demonstrate that the proposed method outperforms the relative state-of-the-art methods on the Shanghai-Tech and XD-Violence datasets and achieves competitive results on the UCF-Crime dataset. This indicates that the proposed method can effectively enhances the performance of anomaly detection, thereby validating its effectiveness in practical applications.

show abstract

Section: Introductionmentioning

confidence: 99%