上海科技大学知识管理系统

Humans detect real-world object anomalies by perceiving, interacting, and reasoning based on object-conditioned phys ical knowledge. The long-term goal of Industrial Anomaly Detection (IAD) is to enable machines to autonomously repli cate this skill. However, current IAD algorithms are largely developed and tested on static, semantically simple datasets, which diverge from real-world scenarios where physical understanding and reasoning are essential. To bridge this gap, we introduce the Physics Anomaly Detection (Phys-AD) dataset, the first large-scale, real-world, physics-grounded video dataset for industrial anomaly detection. Collected using arealrobotarmandmotor, Phys-AD provides a diverse set of dynamic, semantically rich scenarios. The dataset in cludes morethan6400videosacross22real-worldobjectcat egories, interacting with robot arms and motors, and exhibits 47 types of anomalies. Anomaly detection in Phys-AD re quires visual reasoning, combining both physical knowledge andvideocontenttodetermineobjectabnormality. We bench mark state-of-the-art anomaly detection methods under three settings: unsupervisedAD,weakly-supervised AD,and video understanding AD, highlighting their limitations in handling physics-grounded anomalies. Additionally, we introduce the Physics Anomaly Explanation (PAEval) metric, designed to assess the ability of visual-language foundation models to not only detect anomalies but also provide accurate explanations for their underlying physical causes. Our project is available at https://guyao2023.github.io/Phys-AD/.