mlsanityData to 14 Oct 2026

Exploring Plain ViT Reconstruction for Multi-class Unsupervised Anomaly Detection

arXiv 2312.07495 ↗First posted 12 Dec 2023cs.CVcited by 47

Not on the map. Why?

Abstract (arXiv)

This work studies a challenging and practical issue known as multi-class unsupervised anomaly detection (MUAD). This problem requires only normal images for training while simultaneously testing both normal and anomaly images across multiple classes. Existing reconstruction-based methods typically adopt pyramidal networks as encoders and decoders to obtain multi-resolution features, often involving complex sub-modules with extensive handcraft engineering. In contrast, a plain Vision Transformer (ViT) showcasing a more straightforward architecture has proven effective in multiple domains, including detection and segmentation tasks. It is simpler, more effective, and elegant. Following this spirit, we explore the use of only plain ViT features for MUAD. We first abstract a Meta-AD concept by synthesizing current reconstruction-based methods. Subsequently, we instantiate a novel ViT-based ViTAD structure, designed incrementally from both global and local perspectives. This model provide a strong baseline to facilitate future research. Additionally, this paper uncovers several intriguing findings for further investigation. Finally, we comprehensively and fairly benchmark various approaches using eight metrics. Utilizing a basic training regimen with only an MSE loss, ViTAD achieves state-of-the-art results and efficiency on MVTec AD, VisA, and Uni-Medical datasets. Eg, achieving 85.4 mAD that surpasses UniAD by +3.0 for the MVTec AD dataset, and it requires only 1.1 hours and 2.3G GPU memory to complete model training on a single V100 that can serve as a strong baseline to facilitate the development of future research. Full code is available at https://zhangzjn.github.io/projects/ViTAD/.

Key facts. Cited by 47 papers in the map (data through 14 Oct 2026). Builds on 96 of its 118 references that are in the map. First posted on arXiv 12 Dec 2023.

Builds on

96 of 118 references.

References · 96

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  5. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  6. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  7. 2023Segment Anythingcited by 8,234
  8. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  9. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  10. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  11. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  12. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
38 more
  1. 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
  2. 2015Variational Inference with Normalizing Flowscited by 2,170
  3. 2013Describing Textures in the Wildcited by 2,125
  4. 2021Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutionscited by 1,962
  5. 2020Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformerscited by 1,369
  6. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,323
  7. 2021ViLT: Vision-and-Language Transformer Without Convolution or Region Supervisioncited by 1,230
  8. 2021Swin Transformer V2: Scaling Up Capacity and Resolutioncited by 1,228
  9. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  10. 2022Exploring Plain Vision Transformer Backbones for Object Detectioncited by 724
  11. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 594
  12. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  13. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 564
  14. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  15. 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 514
  16. 2022ViTPose: Simple Vision Transformer Baselines for Human Pose Estimationcited by 456
  17. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 421
  18. 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
  19. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  20. 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
  21. 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 321
  22. 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 315
  23. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 300
  24. 2022A Unified Model for Multi-class Anomaly Detectioncited by 268
  25. 2018Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoderscited by 264
  26. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  27. 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 227
  28. 2021VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localizationcited by 223
  29. 2020Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flowscited by 212
  30. 2023SegGPT: Segmenting Everything In Contextcited by 194
  31. 2023Towards Open Vocabulary Learning: A Surveycited by 166
  32. 2023EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latenciescited by 160
  33. 2022DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detectioncited by 154
  34. 2023Deep Industrial Image Anomaly Detection: A Surveycited by 148
  35. 2021Student-Teacher Feature Pyramid Matching for Anomaly Detectioncited by 146
  36. 2023Multimodal Industrial Anomaly Detection via Hybrid Fusioncited by 139
  37. 2023APRIL-GAN: A Zero-/Few-Shot Anomaly Classification and Segmentation Method for CVPR 2023 VAND Workshop Challenge Tracks 1&2: 1st Place on Zero-shot AD and 4th Place on Few-shot ADcited by 130
  38. 2020Explainable Deep One-Class Classificationcited by 127

50 most cited of 96 All 96 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 18; 2025: 14; 2026: 15.

47 papers.

Sort:

By area:

  1. 2024Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detectioncited by 139
  2. 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 127
  3. 2024MambaAD: Exploring State Space Models for Multi-class Unsupervised Anomaly Detectioncited by 113
  4. 2024Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detectioncited by 93
  5. 2024A Survey on Visual Anomaly Detection: Challenge, Approach, and Prospectcited by 69
  6. 2024M3DM-NR: RGB-3D Noisy-Resistant Industrial Anomaly Detection via Multimodal Denoisingcited by 24
  7. 2024Learning Feature Inversion for Multi-class Anomaly Detection under General-purpose COCO-AD Benchmarkcited by 16
  8. 2024Salvaging the Overlooked: Leveraging Class-Aware Contrastive Learning for Multi-Class Anomaly Detectioncited by 15
  9. 2025Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detectioncited by 11
  10. 2024Learning Multi-view Anomaly Detection with Efficient Adaptive Selectioncited by 9
  11. 2024RAD: A Comprehensive Dataset for Benchmarking the Robustness of Image Anomaly Detectioncited by 8
  12. 2025Foundation Models and Transformers for Anomaly Detection: A Surveycited by 8
35 more
  1. 2024Dual-path Frequency Discriminators for Few-shot Anomaly Detectioncited by 7
  2. 2024A Comprehensive Library for Benchmarking Multi-class Visual Anomaly Detectioncited by 7
  3. 2025Towards Accurate Unified Anomaly Segmentationcited by 7
  4. 2025INP-Former++: Advancing Universal Anomaly Detection via Intrinsic Normal Prototypes and Residual Learningcited by 7
  5. 2024Variational Autoencoder for Anomaly Detection: A Comparative Studycited by 6
  6. 2024DMAD: Dual Memory Bank for Real-World Anomaly Detectioncited by 5
  7. 2024Small Object Few-shot Segmentation for Vision-based Industrial Inspectioncited by 5
  8. 2024VQ-Flow: Taming Normalizing Flows for Multi-Class Anomaly Detection via Hierarchical Vector Quantizationcited by 5
  9. 2024ROADS: Robust Prompt-driven Multi-Class Anomaly Detection under Domain Shiftcited by 5
  10. 2025Robust Distribution Alignment for Industrial Anomaly Detection under Distribution Shiftcited by 5
  11. 2026Towards an Incremental Unified Multimodal Anomaly Detection: Augmenting Multimodal Denoising From an Information Bottleneck Perspectivecited by 4
  12. 2024MiniMaxAD: A Lightweight Autoencoder for Feature-Rich Anomaly Detectioncited by 3
  13. 2025VLMDiff: Leveraging Vision-Language Models for Multi-Class Anomaly Detection with Diffusioncited by 3
  14. 2026Is Task-Specific Training Necessary for Anomaly Detection?cited by 2
  15. 2026RAID: Retrieval-Augmented Anomaly Detectioncited by 2
  16. 2026Flow Mismatching: Unsupervised Anomaly Detection via Velocity Discrepancies in Flow Matching Modelscited by 2
  17. 2025Search is All You Need for Few-shot Anomaly Detectioncited by 1
  18. 2025HSS-IAD: A Heterogeneous Same-Sort Industrial Anomaly Detection Datasetcited by 1
  19. 2025Toward Long-Tailed Online Anomaly Detection through Class-Agnostic Conceptscited by 1
  20. 2026Training Free Zero-Shot Visual Anomaly Localization via Diffusion Inversioncited by 1
  21. 2026IMMoE: Incomplete Multi-View Anomaly Detection via Mixture of View Experts Fusioncited by 1
  22. 2025Domain Adaptation of Attention Heads for Zero-shot Anomaly Detectioncited by 0
  23. 2025Crucial-Diff: A Unified Diffusion Model for Crucial Image and Annotation Synthesis in Data-scarce Scenarioscited by 0
  24. 2025Bridge Feature Matching and Cross-Modal Alignment with Mutual-filtering for Zero-shot Anomaly Detectioncited by 0
  25. 2025ShortcutBreaker: Low-Rank Noisy Bottleneck and Frequency Filtering Block for Multi-Class Unsupervised Anomaly Detectioncited by 0
  26. 2025Collaborative Reconstruction and Repair for Multi-class Industrial Anomaly Detectioncited by 0
  27. 2026StructCore: Structure-Aware Image-Level Scoring for Training-Free Unsupervised Anomaly Detectioncited by 0
  28. 2026PDD: Manifold-Prior Diverse Distillation for Medical Anomaly Detectioncited by 0
  29. 2026GroundingAnomaly: Spatially-Grounded Diffusion for Few-Shot Anomaly Synthesiscited by 0
  30. 2026IPAD-CLIP: Teaching CLIP to Detect Image Local Perceptual Artifactscited by 0
  31. 2026Dual Prototype-Conditioned Diffusion Model for Scalable Multi-Class Unsupervised Anomaly Detection in Large Category Spacescited by 0
  32. 2026MambaADv2: Evolving Duality-enhanced State Space Model for Unsupervised Anomaly Detectioncited by 0
  33. 2026MATCH: Flow Matching for Multi-View Anomaly Detectioncited by 0
  34. 2026ArcAD: Anomaly-Rectified Calibration for Cold-Start Supervised Anomaly Detectioncited by 0
  35. 2026ReFP-AD: Rectified Flow Preconditioning for Energy-Based Anomaly Detectioncited by 0