Interactive Anomaly Detection for Articulated Objects via Motion Anticipation

Ankan Kumar Bhunia, Changjian Li, Hakan Bilen

Keywords: articulated objects, anomaly detection, active vision

TL;DR: An interactive anomaly detection method for articulated objects

Abstract: This paper presents a novel problem, interactive anomaly detection (AD) for articulated objects, and introduces a tailored solution that detects functional anomalies by integrating vision, interaction, and anticipation. Unlike traditional AD methods that rely on passive visual observations, our approach actively manipulates objects to reveal anomalies that would otherwise remain hidden. Our method learns to generate a sequence of actions to interact exclusively with normal objects and to anticipate the resulting normal motion. During inference, the model applies predicted actions to the object and compares the observed motion with the anticipated motion to detect anomalies. Additionally, we introduce a new benchmark, PartNet-IAD, for interactive AD, which includes articulated objects with realistic functional anomalies. Experiments show strong generalization to detect anomalies in both seen and unseen object categories. Code and dataset will be released.

Primary Area: Applications (e.g., vision, language, speech and audio, Creative AI)

Submission Number: 12108