Learning Topology-Aware Representations via Test-Time Adaptation for Anomaly Segmentation

2026-06-26

AI Summary

This work addresses the limitations of existing test-time adaptation methods for anomaly segmentation, which rely on pixel-level heuristics that struggle to preserve structural consistency in complex defects and neglect higher-order spatial relationships. To overcome these issues, the study introduces topological data analysis into the test-time adaptation framework for the first time. By applying persistent homology to perform multi-scale cubical complex filtering on anomaly score maps, the method generates connectivity-preserving topological pseudo-labels that guide a lightweight classifier to refine segmentation outputs—without requiring retraining of the backbone model. This approach eliminates dependence on handcrafted thresholds and seamlessly supports both 2D and 3D modalities. Evaluated across six standard benchmarks, it achieves an average F1-score improvement of 15%, with particularly pronounced gains on anomalies exhibiting intricate geometric or structural complexity.

Abstract

Test-time adaptation (TTA) has emerged as a promising paradigm for mitigating distribution shifts in deep models. However, existing TTA approaches for anomaly segmentation remain limited by their reliance on pixel-level heuristics, such as confidence thresholding or entropy minimisation, which fail to preserve structural consistency under noise and texture variation. Moreover, they typically treat anomaly maps as flat intensity fields, ignoring the higher-order spatial relationships that characterise complex defect geometries. We introduce TopoTTA (Topological Test-Time Adaptation), a novel framework that integrates persistent homology, a tool from topological data analysis, into the TTA pipeline to enforce geometric and structural coherence during adaptation. By applying multi-level cubical complex filtration to anomaly score maps, TopoTTA derives robust topological pseudo-labels that guide a lightweight test-time classifier, enhancing segmentation quality without retraining the backbone model. The approach avoids reliance on method-specific raw-score thresholding for mask binarisation, preserves connectivity, and generalises across both 2D and 3D modalities. Extensive experiments across six standard benchmarks (MVTec AD, VisA, Real-IAD, MVTec 3D-AD, AnomalyShapeNet, and MVTec LOCO) demonstrate an average 15% F1 improvement over state-of-the-art unsupervised anomaly detection and segmentation methods, with the largest gains on anomalies exhibiting complex geometric or structural variations. These findings suggest that integrating topological reasoning into test-time adaptation provides a principled route to structure-aware generalisation, bridging the gap between geometric learning and robust adaptation.

Problem

• anomaly segmentation
• test-time adaptation
• structural consistency
• topological data analysis
• distribution shift

Innovation

• Test-Time Adaptation
• Persistent Homology
• Anomaly Segmentation
• Topological Data Analysis
• Structural Coherence