Breaking the Rigid Prior: Towards Articulated 3D Anomaly Detection

TL;DR

This paper introduces ArtiAD, a large-scale benchmark for articulated 3D anomaly detection, and proposes SPA-SDF, a pose-aware implicit model that effectively distinguishes structural defects from pose variations.

Contribution

The paper presents the first benchmark for articulated 3D anomaly detection and a novel shape-pose-aware implicit model that outperforms rigid-based methods.

Findings

SPA-SDF achieves 0.884 object-level AUROC on seen configurations.

SPA-SDF achieves 0.874 object-level AUROC on unseen configurations.

The benchmark includes dense joint-angle variations and multiple anomaly types.

Abstract

Existing 3D anomaly detection methods are built on a rigid prior: normal geometry is pose-invariant and can be canonicalized through registration or alignment. This prior does not hold for articulated objects with hinge or sliding joints, where valid pose changes induce structured geometric variations that cannot be collapsed to a single canonical template, causing pose-induced deformations to be misidentified as anomalies while true structural defects are obscured. No existing benchmark addresses this challenge. We introduce ArtiAD, the first large-scale benchmark for articulated 3D anomaly detection, comprising 15,229 point clouds across 39 object categories with dense joint-angle variations and six structural anomaly types. Each sample is annotated with its joint configuration and part-level motion labels, enabling explicit disentanglement of pose-induced geometry from structural defects. ArtiAD also provides a seen/unseen articulation split to evaluate both interpolation and extrapolation to novel joint configurations. We propose Shape-Pose-Aware Signed Distance Field (SPA-SDF), a baseline that replaces the rigid prior with a continuous pose-conditioned implicit field, factorized into an articulation-independent structural prior and a Fourier-encoded joint embedding. At inference, the articulation state is recovered by minimizing reconstruction energy, and anomalies are identified as point-wise deviations from the learned manifold. SPA-SDF achieves 0.884 object-level AUROC on seen configurations and 0.874 on unseen configurations, substantially outperforming all rigid-based baselines. Our code and benchmark will be publicly released to facilitate future research.