Odd-One-Out: Anomaly Detection by Comparing with Neighbors

Posed multi-view RGB images. Task: To find odd-looking instance(s) in the object set 𝒐 ! ,𝒐 " , 𝒐 # ,𝒐 $ . Fine-grained cross-instance matching 𝒐 ! 𝒐 " 𝒐 % … 𝒐 " 𝒐 # 𝒐 % 𝒐 ! anomaly normal 𝒐 # Object-centric prediction (ambiguous; when definition of 'normality' is unknown) Multi-object AD setting (ours) Standard AD setting Training on seen categories Testing on seen categories Testing on unseen categories Generalization capability on unseen shapes (a) (b) (c) Figure 1. (a) We propose a new anomaly detection task focused on identifying 'odd-looking' objects relative to other instances within a scene. Inspired by real-world quality control in production environments, this task aims to detect subtle variations in geometry and texture, including defects like cracks and fractures, in a group of manufactured samples. (b) Our setting is scene-specific, requiring a comparison of object instances within the input scene, unlike the standard AD setting, which takes only a single object as input. (c) Our matching-based paradigm enables cross-category performance.