Towards Video Anomaly Detection from Event Streams: A Baseline and Benchmark Datasets
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908914434441216 |
|---|---|
| author | Wu, Peng Yan, Yuting Pang, Guansong Sun, Yujia Yan, Qingsen Wang, Peng Zhang, Yanning |
| author_facet | Wu, Peng Yan, Yuting Pang, Guansong Sun, Yujia Yan, Qingsen Wang, Peng Zhang, Yanning |
| contents | Event-based vision, characterized by low redundancy, focus on dynamic motion, and inherent privacy-preserving properties, naturally fits the demands of video anomaly detection (VAD). However, the absence of dedicated event-stream anomaly detection datasets and effective modeling strategies has significantly hindered progress in this field. In this work, we take the first major step toward establishing event-based VAD as a unified research direction. We first construct multiple event-stream based benchmarks for video anomaly detection, featuring synchronized event and RGB recordings. Leveraging the unique properties of events, we then propose an EVent-centric spatiotemporal Video Anomaly Detection framework, namely EWAD, with three key innovations: an event density aware dynamic sampling strategy to select temporally informative segments; a density-modulated temporal modeling approach that captures contextual relations from sparse event streams; and an RGB-to-event knowledge distillation mechanism to enhance event-based representations under weak supervision. Extensive experiments on three benchmarks demonstrate that our EWAD achieves significant improvements over existing approaches, highlighting the potential and effectiveness of event-driven modeling for video anomaly detection. The benchmark datasets will be made publicly available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_24991 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Towards Video Anomaly Detection from Event Streams: A Baseline and Benchmark Datasets Wu, Peng Yan, Yuting Pang, Guansong Sun, Yujia Yan, Qingsen Wang, Peng Zhang, Yanning Computer Vision and Pattern Recognition Event-based vision, characterized by low redundancy, focus on dynamic motion, and inherent privacy-preserving properties, naturally fits the demands of video anomaly detection (VAD). However, the absence of dedicated event-stream anomaly detection datasets and effective modeling strategies has significantly hindered progress in this field. In this work, we take the first major step toward establishing event-based VAD as a unified research direction. We first construct multiple event-stream based benchmarks for video anomaly detection, featuring synchronized event and RGB recordings. Leveraging the unique properties of events, we then propose an EVent-centric spatiotemporal Video Anomaly Detection framework, namely EWAD, with three key innovations: an event density aware dynamic sampling strategy to select temporally informative segments; a density-modulated temporal modeling approach that captures contextual relations from sparse event streams; and an RGB-to-event knowledge distillation mechanism to enhance event-based representations under weak supervision. Extensive experiments on three benchmarks demonstrate that our EWAD achieves significant improvements over existing approaches, highlighting the potential and effectiveness of event-driven modeling for video anomaly detection. The benchmark datasets will be made publicly available. |
| title | Towards Video Anomaly Detection from Event Streams: A Baseline and Benchmark Datasets |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.24991 |