Open-Vocabulary Video Anomaly Detection

arXiv:2311.07042v3 Announce Type: replace
Abstract: Video anomaly detection (VAD) with weak supervision has achieved remarkable performance in utilizing video-level labels to discriminate whether a video frame is normal or abnormal. However, current approaches are inherently limited to a closed-set setting and may struggle in open-world applications where there can be anomaly categories in the test data unseen during training. A few recent studies attempt to tackle a more realistic setting, open-set VAD, which aims to detect unseen anomalies given seen …

abstract anomaly anomaly detection applications arxiv cs.cv current data detection however labels normal open-world performance set struggle supervision test type video world