mlsanityData to 14 Oct 2026

Visual Anomaly Detection under Complex View-Illumination Interplay: A Large-Scale Benchmark

arXiv 2505.10996 ↗First posted 16 May 2025cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

The practical deployment of Visual Anomaly Detection (VAD) systems is hindered by their sensitivity to real-world imaging variations, particularly the complex interplay between viewpoint and illumination which drastically alters defect visibility. Current benchmarks largely overlook this critical challenge. We introduce Multi-View Multi-Illumination Anomaly Detection (M2AD), a new large-scale benchmark comprising 119,880 high-resolution images designed explicitly to probe VAD robustness under such interacting conditions. By systematically capturing 999 specimens across 10 categories using 12 synchronized views and 10 illumination settings (120 configurations total), M2AD enables rigorous evaluation. We establish two evaluation protocols: M2AD-Synergy tests the ability to fuse information across diverse configurations, and M2AD-Invariant measures single-image robustness against realistic view-illumination effects. Our extensive benchmarking shows that state-of-the-art VAD methods struggle significantly on M2AD, demonstrating the profound challenge posed by view-illumination interplay. This benchmark serves as an essential tool for developing and validating VAD methods capable of overcoming real-world complexities. Our full dataset and test suite will be released at https://hustcyq.github.io/M2AD to facilitate the field.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 34 of its 39 references that are in the map. First posted on arXiv 16 May 2025.

Builds on

34 of 39 references.

References · 34

  1. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  2. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  3. 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 271
  4. 2021VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localizationcited by 223
  5. 2022Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstructioncited by 212
  6. 2021The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localizationcited by 178
  7. 2023Multimodal Industrial Anomaly Detection via Hybrid Fusioncited by 139
  8. 2024Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detectioncited by 139
  9. 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 127
  10. 2024MambaAD: Exploring State Space Models for Multi-class Unsupervised Anomaly Detectioncited by 113
  11. 2022IRON: Inverse Rendering by Optimizing Neural SDFs and Materials from Photometric Imagescited by 111
  12. 2021Mixed supervision for surface-defect detection: from weakly to fully supervised learningcited by 101
22 more
  1. 2024Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detectioncited by 93
  2. 2023Real3D-AD: A Dataset of Point Cloud Anomaly Detectioncited by 87
  3. 2024A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localizationcited by 74
  4. 2024Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularitycited by 70
  5. 2022The Eyecandies Dataset for Unsupervised Multimodal Anomaly Detection and Localizationcited by 57
  6. 2023Collaborative Discrepancy Optimization for Reliable Image Anomaly Localizationcited by 57
  7. 2025Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detectioncited by 48
  8. 2023PAD: A Dataset and Benchmark for Pose-agnostic Anomaly Detectioncited by 46
  9. 2023EasyNet: An Easy Network for 3D Industrial Anomaly Detectioncited by 43
  10. 2025The MVTec AD 2 Dataset: Advanced Scenarios for Unsupervised Anomaly Detectioncited by 30
  11. 2024Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detectioncited by 28
  12. 2024MANTA: A Large-Scale Multi-View and Visual-Text Anomaly Detection Dataset for Tiny Objectscited by 26
  13. 2024Multi-Sensor Object Anomaly Detection: Unifying Appearance, Geometry, and Internal Propertiescited by 26
  14. 2023MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detectioncited by 25
  15. 2024SplatPose & Detect: Pose-Agnostic 3D Anomaly Detectioncited by 25
  16. 2023Industrial Anomaly Detection with Domain Shift: A Real-world Dataset and Masked Multi-scale Reconstructioncited by 23
  17. 2025Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detectioncited by 12
  18. 2025Towards Visual Discrimination and Reasoning of Real-World Physical Dynamics: Physics-Grounded Anomaly Detectioncited by 10
  19. 2024Learning Multi-view Anomaly Detection with Efficient Adaptive Selectioncited by 9
  20. 2024VarAD: Lightweight High-Resolution Image Anomaly Detection via Visual Autoregressive Modelingcited by 9
  21. 2024RAD: A Dataset and Benchmark for Real-Life Anomaly Detection with Robotic Observationscited by 5
  22. 2024Attention Fusion Reverse Distillation for Multi-Lighting Image Anomaly Detectioncited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 3.

4 papers.

  1. 2025A Comprehensive Survey for Real-World Industrial Defect Detection: Challenges, Approaches, and Prospectscited by 2
  2. 2026PiCo: Active Manifold Canonicalization for Robust Robotic Visual Anomaly Detectioncited by 0
  3. 2026MATCH: Flow Matching for Multi-View Anomaly Detectioncited by 0
  4. 2026FreSia: Frequency-Semantic Instantiation and Alignment for Multivariate Time Series Analysiscited by 0