Weakly supervised video anomaly detection and localization with spatio-temporal prompts

• Main Authors: WU, Peng, ZHOU, Xuerong, PANG, Guansong, YANG, Zhiwei, YAN, Qingsen, WANG, Peng, ZHANG, Yanning
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2026
• Subjects: Scene anomaly detection; Visual content-based indexing and retrieval; Video anomaly detection; Spatio-temporal detection; Language-image pre-training; Artificial Intelligence and Robotics; Graphics and Human Computer Interfaces
• Online Access: https://ink.library.smu.edu.sg/sis_research/9758; https://ink.library.smu.edu.sg/context/sis_research/article/10758/viewcontent/2408.05905v2.pdf
• Institution: Singapore Management University

Current weakly supervised video anomaly detection (WSVAD) task aims to achieve frame-level anomalous event detection with only coarse video-level annotations available. Existing works typically involve extracting global features from full-resolution video frames and training frame-level classifiers...