mlsanityData to 14 Oct 2026

INP-Former++: Advancing Universal Anomaly Detection via Intrinsic Normal Prototypes and Residual Learning

arXiv 2506.03660 ↗First posted 4 Jun 2025cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

Anomaly detection (AD) is essential for industrial inspection and medical diagnosis, yet existing methods typically rely on ``comparing'' test images to normal references from a training set. However, variations in appearance and positioning often complicate the alignment of these references with the test image, limiting detection accuracy. We observe that most anomalies manifest as local variations, meaning that even within anomalous images, valuable normal information remains. We argue that this information is useful and may be more aligned with the anomalies since both the anomalies and the normal information originate from the same image. Therefore, rather than relying on external normality from the training set, we propose INP-Former, a novel method that extracts Intrinsic Normal Prototypes (INPs) directly from the test image. Specifically, we introduce the INP Extractor, which linearly combines normal tokens to represent INPs. We further propose an INP Coherence Loss to ensure INPs can faithfully represent normality for the testing image. These INPs then guide the INP-guided Decoder to reconstruct only normal tokens, with reconstruction errors serving as anomaly scores. Additionally, we propose a Soft Mining Loss to prioritize hard-to-optimize samples during training. INP-Former achieves state-of-the-art performance in single-class, multi-class, and few-shot AD tasks across MVTec-AD, VisA, and Real-IAD, positioning it as a versatile and universal solution for AD. Remarkably, INP-Former also demonstrates some zero-shot AD capability. Furthermore, we propose a soft version of the INP Coherence Loss and enhance INP-Former by incorporating residual learning, leading to the development of INP-Former++. The proposed method significantly improves detection performance across single-class, multi-class, semi-supervised, few-shot, and zero-shot settings.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 57 of its 76 references that are in the map. First posted on arXiv 4 Jun 2025.

Builds on

57 of 76 references.

References · 57

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,718
  5. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  6. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  7. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 594
  8. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 564
  9. 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 514
  10. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 421
  11. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  12. 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 315
38 more
  1. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 300
  2. 2019Deep Semi-Supervised Anomaly Detectioncited by 291
  3. 2020Learning Memory-guided Normality for Anomaly Detectioncited by 278
  4. 2022A Unified Model for Multi-class Anomaly Detectioncited by 268
  5. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  6. 2021Adaptive Prototype Learning and Allocation for Few-Shot Segmentationcited by 207
  7. 2022DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detectioncited by 154
  8. 2024Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detectioncited by 139
  9. 2022Registration based Few-Shot Anomaly Detectioncited by 133
  10. 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
  11. 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 127
  12. 2024RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detectioncited by 116
  13. 2024MambaAD: Exploring State Space Models for Multi-class Unsupervised Anomaly Detectioncited by 113
  14. 2024PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detectioncited by 98
  15. 2024Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detectioncited by 93
  16. 2024Toward Generalist Anomaly Detection via In-context Residual Learning with Few-shot Sample Promptscited by 89
  17. 2023Stable and low-precision training for large-scale vision-language modelscited by 75
  18. 2024A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localizationcited by 74
  19. 2022Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detectioncited by 72
  20. 2021Learning Normal Dynamics in Videos with Meta Prototype Networkcited by 71
  21. 2021Explainable Deep Few-shot Anomaly Detection with Deviation Networkscited by 69
  22. 2024A Survey on Visual Anomaly Detection: Challenge, Approach, and Prospectcited by 69
  23. 2024Adapting Visual-Language Models for Generalizable Anomaly Detection in Medical Imagescited by 69
  24. 2023Hierarchical Vector Quantized Transformer for Multi-class Unsupervised Anomaly Detectioncited by 66
  25. 2023IM-IAD: Industrial Image Anomaly Detection Benchmark in Manufacturingcited by 64
  26. 2023ReContrast: Domain-Specific Anomaly Detection via Contrastive Reconstructioncited by 63
  27. 2023Collaborative Discrepancy Optimization for Reliable Image Anomaly Localizationcited by 57
  28. 2022Prototypical Residual Networks for Anomaly Detection and Localizationcited by 56
  29. 2024SoftPatch: Unsupervised Anomaly Detection with Noisy Datacited by 55
  30. 2025Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detectioncited by 48
  31. 2023Exploring Plain ViT Reconstruction for Multi-class Unsupervised Anomaly Detectioncited by 47
  32. 2020Rethinking Assumptions in Deep Anomaly Detectioncited by 45
  33. 2019Learn to Segment Retinal Lesions and Beyondcited by 33
  34. 2023ReConPatch : Contrastive Patch Representation Learning for Industrial Anomaly Detectioncited by 30
  35. 2024Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detectioncited by 28
  36. 2024ResAD: A Simple Framework for Class Generalizable Anomaly Detectioncited by 25
  37. 2024Prior Normality Prompt Transformer for Multi-class Industrial Image Anomaly Detectioncited by 19
  38. 2025Learning to Detect Multi-class Anomalies with Just One Normal Image Promptcited by 18

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 3.

7 papers.

  1. 2025A Comprehensive Survey for Real-World Industrial Defect Detection: Challenges, Approaches, and Prospectscited by 2
  2. 2025ResAD++: Towards Class Agnostic Anomaly Detection via Residual Feature Learningcited by 2
  3. 2025CoPS: Conditional Prompt Synthesis for Zero-Shot Anomaly Detectioncited by 1
  4. 2025AnyAD: Unified Any-Modality Anomaly Detection in Incomplete Multi-Sequence MRIcited by 1
  5. 2026ReFP-AD: Rectified Flow Preconditioning for Energy-Based Anomaly Detectioncited by 0
  6. 2026PADFormer: Pose-agnostic Anomaly Detection from Sparse View Imagescited by 0
  7. 2026What Remains Normal? Clean Images Miss Useful Near-Defect Normal Patches for Anomaly Detectioncited by 0