mlsanityData to 14 Oct 2026

Fully Convolutional Cross-Scale-Flows for Image-based Defect Detection

arXiv 2110.02855 ↗First posted 6 Oct 2021cs.CVcited by 117

Not on the map. Why?

Abstract (arXiv)

In industrial manufacturing processes, errors frequently occur at unpredictable times and in unknown manifestations. We tackle the problem of automatic defect detection without requiring any image samples of defective parts. Recent works model the distribution of defect-free image data, using either strong statistical priors or overly simplified data representations. In contrast, our approach handles fine-grained representations incorporating the global and local image context while flexibly estimating the density. To this end, we propose a novel fully convolutional cross-scale normalizing flow (CS-Flow) that jointly processes multiple feature maps of different scales. Using normalizing flows to assign meaningful likelihoods to input samples allows for efficient defect detection on image-level. Moreover, due to the preserved spatial arrangement the latent space of the normalizing flow is interpretable which enables to localize defective regions in the image. Our work sets a new state-of-the-art in image-level defect detection on the benchmark datasets Magnetic Tile Defects and MVTec AD showing a 100% AUROC on 4 out of 15 classes.

Key facts. Cited by 117 papers in the map (data through 14 Oct 2026). Builds on 27 of its 42 references that are in the map. First posted on arXiv 6 Oct 2021.

Builds on

27 of 42 references.

References · 27

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  4. 2015Variational Inference with Normalizing Flowscited by 2,170
  5. 2016Density estimation using Real NVPcited by 2,042
  6. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  7. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 640
  8. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 594
  9. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 564
  10. 2020CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instancescited by 441
  11. 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
  12. 2015MADE: Masked Autoencoder for Distribution Estimationcited by 307
15 more
  1. 2018Deep Anomaly Detection Using Geometric Transformationscited by 269
  2. 2018Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoderscited by 264
  3. 2020Why Normalizing Flows Fail to Detect Out-of-Distribution Datacited by 221
  4. 2020Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flowscited by 212
  5. 2019Guided Image Generation with Conditional Invertible Neural Networkscited by 147
  6. 2020Explainable Deep One-Class Classificationcited by 127
  7. 2020Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detectioncited by 115
  8. 2016Deep Structured Energy Based Models for Anomaly Detectioncited by 114
  9. 2021Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flowscited by 85
  10. 2020TOAD-GAN: Coherent Style Level Generation from a Single Examplecited by 22
  11. 2018Are pre-trained CNNs good feature extractors for anomaly detection in surveillance videos?cited by 17
  12. 2019Structuring Autoencoderscited by 11
  13. 2020Anomaly Detection in Trajectory Data with Normalizing Flowscited by 11
  14. 2019Normalizing flows for novelty detection in industrial time series datacited by 6
  15. 2018Video Event Recognition and Anomaly Detection by Combining Gaussian Process and Hierarchical Dirichlet Process Modelscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 14; 2023: 33; 2024: 27; 2025: 28; 2026: 15.

117 papers.

Sort:

By area:

  1. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  2. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  3. 2023EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latenciescited by 160
  4. 2023Deep Industrial Image Anomaly Detection: A Surveycited by 148
  5. 2024Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detectioncited by 139
  6. 2022Asymmetric Student-Teacher Networks for Industrial Anomaly Detectioncited by 117
  7. 2023BMAD: Benchmarks for Medical Anomaly Detectioncited by 82
  8. 2023PyramidFlow: High-Resolution Defect Contrastive Localization using Pyramid Normalizing Flowcited by 74
  9. 2022Deep Learning for Unsupervised Anomaly Localization in Industrial Images: A Surveycited by 68
  10. 2023IM-IAD: Industrial Image Anomaly Detection Benchmark in Manufacturingcited by 64
  11. 2022Prototypical Residual Networks for Anomaly Detection and Localizationcited by 56
  12. 2024AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2cited by 53
38 more
  1. 2023DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detectioncited by 46
  2. 2023PAD: A Dataset and Benchmark for Pose-agnostic Anomaly Detectioncited by 46
  3. 2023TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detectioncited by 43
  4. 2023Cheating Depth: Enhancing 3D Surface Anomaly Detection via Depth Simulationcited by 33
  5. 2023Focus the Discrepancy: Intra- and Inter-Correlation Learning for Image Anomaly Detectioncited by 32
  6. 2024Unsupervised Continual Anomaly Detection with Contrastively-learned Promptcited by 31
  7. 2025The MVTec AD 2 Dataset: Advanced Scenarios for Unsupervised Anomaly Detectioncited by 30
  8. 2025AdaptCLIP: Adapting CLIP for Universal Visual Anomaly Detectioncited by 29
  9. 2024Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detectioncited by 28
  10. 2023MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detectioncited by 25
  11. 2024SplatPose & Detect: Pose-Agnostic 3D Anomaly Detectioncited by 25
  12. 2024ResAD: A Simple Framework for Class Generalizable Anomaly Detectioncited by 25
  13. 2023Component-aware anomaly detection framework for adjustable and logical industrial visual inspectioncited by 22
  14. 2022A Survey on Unsupervised Anomaly Detection Algorithms for Industrial Imagescited by 21
  15. 2024Learning Unified Reference Representation for Unsupervised Multi-class Anomaly Detectioncited by 19
  16. 2022Reconstruction from edge image combined with color and gradient difference for industrial surface anomaly detectioncited by 18
  17. 2025Learning to Detect Multi-class Anomalies with Just One Normal Image Promptcited by 18
  18. 2024Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detectioncited by 17
  19. 2022U-Flow: A U-shaped Normalizing Flow for Anomaly Detection with Unsupervised Thresholdcited by 16
  20. 2023InsPLAD: A Dataset and Benchmark for Power Line Asset Inspection in UAV Imagescited by 15
  21. 2024Kernel-Aware Graph Prompt Learning for Few-Shot Anomaly Detectioncited by 15
  22. 2025MetaUAS: Universal Anomaly Segmentation with One-Prompt Meta-Learningcited by 13
  23. 2024Revisiting Multimodal Fusion for 3D Anomaly Detection from an Architectural Perspectivecited by 12
  24. 2022Benchmarking Unsupervised Anomaly Detection and Localizationcited by 11
  25. 2022Data Augmentation is a Hyperparameter: Cherry-picked Self-Supervision for Unsupervised Anomaly Detection is Creating the Illusion of Successcited by 11
  26. 2024Text-Guided Variational Image Generation for Industrial Anomaly Detection and Segmentationcited by 11
  27. 2025Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detectioncited by 11
  28. 2026VisualAD: Language-Free Zero-Shot Anomaly Detection via Vision Transformercited by 11
  29. 2023REB: Reducing Biases in Representation for Industrial Anomaly Detectioncited by 10
  30. 2023What makes a good data augmentation for few-shot unsupervised image anomaly detection?cited by 8
  31. 2024FUN-AD: Fully Unsupervised Learning for Anomaly Detection with Noisy Training Datacited by 8
  32. 2025Towards Real Unsupervised Anomaly Detection Via Confident Meta-Learningcited by 8
  33. 2022Unsupervised Visual Defect Detection with Score-Based Generative Modelcited by 7
  34. 2023Take 5: Interpretable Image Classification with a Handful of Featurescited by 7
  35. 2023The voraus-AD Dataset for Anomaly Detection in Robot Applicationscited by 7
  36. 2025Towards Accurate Unified Anomaly Segmentationcited by 7
  37. 2025Registration is a Powerful Rotation-Invariance Learner for 3D Anomaly Detectioncited by 7
  38. 2022Invertible Monotone Operators for Normalizing Flowscited by 6

50 most cited of 117 All 117 (JSON)