蒸馏聚合知识用于弱监督视频异常检测
Distilling Aggregated Knowledge for Weakly-Supervised Video Anomaly Detection
AI总结：
提出DAKD方法，通过双层蒸馏将多骨干聚合知识转移至单骨干学生模型，结合解纠缠交叉注意力聚合网络，在三个基准数据集上实现最先进的视频异常检测性能。
摘要：
视频异常检测旨在开发能够识别监控视频中异常事件的自动化模型。该任务的基准设置极具挑战性，原因在于：i) 训练集规模有限，ii) 提供的弱监督仅为视频级标签，以及 iii) 异常事件稀缺导致的内在类别不平衡。在本工作中，我们表明将多个骨干网络的聚合表示蒸馏到单骨干学生模型中，能够实现最先进的性能。具体而言，我们开发了一种双层蒸馏方法，并配合一种新颖的基于解纠缠交叉注意力的特征聚合网络。我们提出的方法 DAKD（使用解纠缠注意力蒸馏聚合知识）在多个基准数据集上展现出优于现有方法的性能。值得注意的是，我们在 UCF-Crime、ShanghaiTech 和 XD-Violence 数据集上分别取得了 1.36%、0.78% 和 7.02% 的显著提升。
英文摘要：
Video anomaly detection aims to develop automated models capable of identifying abnormal events in surveillance videos. The benchmark setup for this task is extremely challenging due to: i) the limited size of the training sets, ii) weak supervision provided in terms of video-level labels, and iii) intrinsic class imbalance induced by the scarcity of abnormal events. In this work, we show that distilling knowledge from aggregated representations of multiple backbones into a single-backbone Student model achieves state-of-the-art performance. In particular, we develop a bi-level distillation approach along with a novel disentangled cross-attention-based feature aggregation network. Our proposed approach, DAKD (Distilling Aggregated Knowledge with Disentangled Attention), demonstrates superior performance compared to existing methods across multiple benchmark datasets. Notably, we achieve significant improvements of 1.36%, 0.78%, and 7.02% on the UCF-Crime, ShanghaiTech, and XD-Violence datasets, respectively.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
