mlsanityData to 14 Oct 2026

TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection

arXiv 2311.09999 ↗First posted 16 Nov 2023cs.CVcited by 43

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Surface anomaly detection is a vital component in manufacturing inspection. Current discriminative methods follow a two-stage architecture composed of a reconstructive network followed by a discriminative network that relies on the reconstruction output. Currently used reconstructive networks often produce poor reconstructions that either still contain anomalies or lack details in anomaly-free regions. Discriminative methods are robust to some reconstructive network failures, suggesting that the discriminative network learns a strong normal appearance signal that the reconstructive networks miss. We reformulate the two-stage architecture into a single-stage iterative process that allows the exchange of information between the reconstruction and localization. We propose a novel transparency-based diffusion process where the transparency of anomalous regions is progressively increased, restoring their normal appearance accurately while maintaining the appearance of anomaly-free regions using localization cues of previous steps. We implement the proposed process as TRANSparency DifFUSION (TransFusion), a novel discriminative anomaly detection method that achieves state-of-the-art performance on both the VisA and the MVTec AD datasets, with an image-level AUROC of 98.5% and 99.2%, respectively. Code: https://github.com/MaticFuc/ECCV_TransFusion

Key facts. Cited by 43 papers in the map (data through 14 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 32 of its 45 references that are in the map. First posted on arXiv 16 Nov 2023.

Builds on

32 of 45 references.

References · 32

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Denoising Diffusion Implicit Modelscited by 9,781
  3. 2023Consistency Modelscited by 1,822
  4. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
  5. 2020DiffWave: A Versatile Diffusion Model for Audio Synthesiscited by 1,354
  6. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,027
  7. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  8. 2022On Distillation of Guided Diffusion Modelscited by 725
  9. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  10. 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 514
  11. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 421
  12. 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
20 more
  1. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  2. 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
  3. 2022DiffusionDet: Diffusion Model for Object Detectioncited by 354
  4. 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 321
  5. 2021SegDiff: Image Segmentation with Diffusion Probabilistic Modelscited by 280
  6. 2022Cold Diffusion: Inverting Arbitrary Image Transforms Without Noisecited by 276
  7. 2018Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoderscited by 264
  8. 2019ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed datacited by 261
  9. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  10. 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 227
  11. 2022MedSegDiff: Medical Image Segmentation with Diffusion Probabilistic Modelcited by 180
  12. 2023EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latenciescited by 160
  13. 2022DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detectioncited by 154
  14. 2022FastDiff: A Fast Conditional Diffusion Model for High-Quality Speech Synthesiscited by 136
  15. 2021Fully Convolutional Cross-Scale-Flows for Image-based Defect Detectioncited by 117
  16. 2022Asymmetric Student-Teacher Networks for Industrial Anomaly Detectioncited by 117
  17. 2022DSR -- A dual subspace re-projection network for surface anomaly detectioncited by 90
  18. 2022MemSeg: A semi-supervised method for image surface defect detection using differences and commonalitiescited by 68
  19. 2022U-Flow: A U-shaped Normalizing Flow for Anomaly Detection with Unsupervised Thresholdcited by 16
  20. 2022N-pad : Neighboring Pixel-based Industrial Anomaly Detectioncited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 4; 2025: 22; 2026: 17.

43 papers.

Sort:

By area:

  1. 2025Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Modelscited by 54
  2. 2025One-for-More: Continual Diffusion Model for Anomaly Detectioncited by 19
  3. 2024SuperSimpleNet: Unifying Unsupervised and Supervised Learning for Fast and Reliable Surface Defect Detectioncited by 12
  4. 2025Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detectioncited by 11
  5. 2025Towards Real Unsupervised Anomaly Detection Via Confident Meta-Learningcited by 8
  6. 2025Registration is a Powerful Rotation-Invariance Learner for 3D Anomaly Detectioncited by 7
  7. 2026SubspaceAD: Training-Free Few-Shot Anomaly Detection via Subspace Modelingcited by 7
  8. 2025A Survey on Foundation-Model-Based Industrial Defect Detectioncited by 5
  9. 2025No Label Left Behind: A Unified Surface Defect Detection Model for all Supervision Regimescited by 4
  10. 2025SALAD -- Semantics-Aware Logical Anomaly Detectioncited by 4
  11. 2026AnomalyVFM -- Transforming Vision Foundation Models into Zero-Shot Anomaly Detectorscited by 4
  12. 2025InvAD: Inversion-based Reconstruction-Free Anomaly Detection with Diffusion Modelscited by 3
31 more
  1. 2025Crane: Context-Guided Prompt Learning and Attention Refinement for Zero-Shot Anomaly Detectioncited by 3
  2. 2025RoBiS: Robust Binary Segmentation for High-Resolution Industrial Imagescited by 3
  3. 2025RDDPM: Robust Denoising Diffusion Probabilistic Model for Unsupervised Anomaly Segmentationcited by 3
  4. 2025VLMDiff: Leveraging Vision-Language Models for Multi-Class Anomaly Detection with Diffusioncited by 3
  5. 2026A Semantically Disentangled Unified Model for Multi-category 3D Anomaly Detectioncited by 3
  6. 2024Friend or Foe? Harnessing Controllable Overfitting for Anomaly Detectioncited by 2
  7. 2025Quality-Aware Language-Conditioned Local Auto-Regressive Anomaly Synthesis and Detectioncited by 2
  8. 2026RAID: Retrieval-Augmented Anomaly Detectioncited by 2
  9. 2026Flow Mismatching: Unsupervised Anomaly Detection via Velocity Discrepancies in Flow Matching Modelscited by 2
  10. 2025Beyond Academic Benchmarks: Critical Analysis and Best Practices for Visual Industrial Anomaly Detectioncited by 1
  11. 2025MathPhys-Guided Coarse-to-Fine Anomaly Synthesis with SQE-Driven Bi-Level Optimization for Anomaly Detectioncited by 1
  12. 20253D-ADAM: A Dataset for 3D Anomaly Detection in Additive Manufacturingcited by 1
  13. 2025Toward Long-Tailed Online Anomaly Detection through Class-Agnostic Conceptscited by 1
  14. 2025ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretrainingcited by 1
  15. 2026Training Free Zero-Shot Visual Anomaly Localization via Diffusion Inversioncited by 1
  16. 2026Multimodal Industrial Anomaly Detection via Geometric Priorcited by 1
  17. 2026Diffuse to Detect: Generative Diffusion Models for Unsupervised IC Anomaly Detectioncited by 1
  18. 2026Normality-Preserving Continual Industrial Anomaly Detection via Orthogonal LoRA Bankscited by 1
  19. 2024Divide and Conquer: High-Resolution Industrial Anomaly Detection via Memory Efficient Tiled Ensemblecited by 0
  20. 2024Tuned Reverse Distillation: Enhancing Multimodal Industrial Anomaly Detection with Crossmodal Tunerscited by 0
  21. 2025Unsupervised Integrated-Circuit Defect Segmentation via Image-Intrinsic Normalitycited by 0
  22. 2025Image-Intrinsic Priors for Integrated Circuit Defect Detection and Novel Class Discovery via Self-Supervised Learningcited by 0
  23. 2025RcAE: Recursive Reconstruction Framework for Unsupervised Industrial Anomaly Detectioncited by 0
  24. 2026Test-Time Adaptation for Anomaly Segmentation via Topology-Aware Optimal Transport Chainingcited by 0
  25. 2026ChangeFlow -- Latent Rectified Flow for Change Detection in Remote Sensingcited by 0
  26. 2026Dual Prototype-Conditioned Diffusion Model for Scalable Multi-Class Unsupervised Anomaly Detection in Large Category Spacescited by 0
  27. 2026Learning Topology-Aware Representations via Test-Time Adaptation for Anomaly Segmentationcited by 0
  28. 2026LogiCo: A Unified Framework for Logical and Structural Anomaly Detectioncited by 0
  29. 2026TC-MAF: Train-Calibrated Bounded Multi-Evidence Fusion for Multimodal Industrial Anomaly Detectioncited by 0
  30. 2026M2P-AD: Memory-to-Prototype Learning with Boundary-aware Score Refinement for 3D Anomaly Detectioncited by 0
  31. 2026DuoAD: Leveraging [CLS] Dual Characteristics for Training-Free Few-Shot Anomaly Detectioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Enhancing Multi-Class Anomaly Detection via Diffusion Refinement with Dual Conditioning
  2. 2023DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection
  3. 2023DiAD: A Diffusion-based Framework for Multi-class Anomaly Detection
  4. 2025Single-Step Reconstruction-Free Anomaly Detection and Segmentation via Diffusion Models
  5. 2023AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Model
  6. 2025HDM: Hybrid Diffusion Model for Unified Image Anomaly Detection
  7. 2024Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generation
  8. 2024R3D-AD: Reconstruction via Diffusion for 3D Anomaly Detection