基于弱监督双编码器模型的监控视频异常事件识别

Recognition of Abnormal Events in Surveillance Videos using Weakly Supervised Dual-Encoder Models

November 17, 2025

作者: Noam Tsfaty, Avishai Weizman, Liav Cohen, Moshe Tshuva, Yehudit Aperstein

cs.AI

摘要

我们提出了一种仅需视频级监督即可检测监控视频中罕见多样化异常的方法。通过双主干网络架构融合卷积与Transformer特征表示，并采用top-k池化策略，在UCF-Crime数据集上实现了90.7%的曲线下面积（AUC）。

English

We address the challenge of detecting rare and diverse anomalies in surveillance videos using only video-level supervision. Our dual-backbone framework combines convolutional and transformer representations through top-k pooling, achieving 90.7% area under the curve (AUC) on the UCF-Crime dataset.