mlsanityData to 14 Oct 2026

Few-Shot Defect Image Generation via Defect-Aware Feature Manipulation

arXiv 2303.02389 ↗First posted 4 Mar 2023cs.CVcited by 67

Not on the map. Why?

Abstract (arXiv)

The performances of defect inspection have been severely hindered by insufficient defect images in industries, which can be alleviated by generating more samples as data augmentation. We propose the first defect image generation method in the challenging few-shot cases. Given just a handful of defect images and relatively more defect-free ones, our goal is to augment the dataset with new defect images. Our method consists of two training stages. First, we train a data-efficient StyleGAN2 on defect-free images as the backbone. Second, we attach defect-aware residual blocks to the backbone, which learn to produce reasonable defect masks and accordingly manipulate the features within the masked regions by training the added modules on limited defect images. Extensive experiments on MVTec AD dataset not only validate the effectiveness of our method in generating realistic and diverse defect images, but also manifest the benefits it brings to downstream defect inspection tasks. Codes are available at https://github.com/Ldhlwh/DFMGAN.

Key facts. Cited by 67 papers in the map (data through 14 Oct 2026). Builds on 23 of its 31 references that are in the map. First posted on arXiv 4 Mar 2023.

Builds on

23 of 31 references.

References · 23

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  3. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  4. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  5. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
  6. 2017Improved Training of Wasserstein GANscited by 2,235
  7. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,628
  8. 2018Demystifying MMD GANscited by 1,293
  9. 2020Training Generative Adversarial Networks with Limited Datacited by 1,186
  10. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,098
  11. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  12. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
11 more
  1. 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
  2. 2020Differentiable Augmentation for Data-Efficient GAN Trainingcited by 363
  3. 2021Few-shot Image Generation via Cross-domain Correspondencecited by 207
  4. 2018Deep Autoencoding Models for Unsupervised Anomaly Segmentation in Brain MR Imagescited by 146
  5. 2019Mode Seeking Generative Adversarial Networks for Diverse Image Synthesiscited by 138
  6. 2019MineGAN: effective knowledge transfer from GANs to target domains with few imagescited by 106
  7. 2019Image Generation From Small Datasets via Batch Statistics Adaptationcited by 102
  8. 2021Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspectioncited by 77
  9. 2020Few-Shot Adaptation of Generative Adversarial Networkscited by 64
  10. 2020On Leveraging Pretrained GANs for Generation with Limited Datacited by 11
  11. 2020Few-Shot Defect Segmentation Leveraging Abundant Normal Training Samples Through Normal Background Regularization and Crop-and-Paste Operationcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 3; 2024: 15; 2025: 31; 2026: 18.

67 papers.

Sort:

By area:

  1. 2024RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detectioncited by 116
  2. 2023AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Modelcited by 111
  3. 2024A Survey on Visual Anomaly Detection: Challenge, Approach, and Prospectcited by 69
  4. 2024Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generationcited by 32
  5. 2024SeaS: Few-shot Industrial Anomaly Image Generation with Separation and Sharing Fine-tuningcited by 27
  6. 2024Unseen Visual Anomaly Generationcited by 23
  7. 2025Noise-Consistent Siamese-Diffusion for Medical Image Synthesis and Segmentationcited by 23
  8. 2024AnomalyXFusion: Multi-modal Anomaly Synthesis with Diffusioncited by 21
  9. 2025Boosting Global-Local Feature Matching via Anomaly Synthesis for Multi-Class Point Cloud Anomaly Detectioncited by 16
  10. 2025DefectFill: Realistic Defect Generation with Inpainting Diffusion Model for Visual Inspectioncited by 13
  11. 2023WeditGAN: Few-Shot Image Generation via Latent Space Relocationcited by 12
  12. 2024Revisiting Multimodal Fusion for 3D Anomaly Detection from an Architectural Perspectivecited by 12
38 more
  1. 2023Image Synthesis under Limited Data: A Survey and Taxonomycited by 9
  2. 2024Few-shot Defect Image Generation based on Consistency Modelingcited by 8
  3. 2025Anomaly Detection and Generation with Diffusion Models: A Surveycited by 8
  4. 2024AnomalyControl: Learning Cross-modal Semantic Features for Controllable Anomaly Synthesiscited by 7
  5. 2025A Survey on Industrial Anomaly Synthesiscited by 7
  6. 2024A Novel Approach to Industrial Defect Generation through Blended Latent Diffusion Model with Online Adaptationcited by 6
  7. 2024Adapted-MoE: Mixture of Experts with Test-Time Adaption for Anomaly Detectioncited by 5
  8. 2025Component-aware Unsupervised Logical Anomaly Generation for Industrial Anomaly Detectioncited by 5
  9. 2025A Survey on Foundation-Model-Based Industrial Defect Detectioncited by 5
  10. 2025AnomalyPainter: Vision-Language-Diffusion Synergy for Zero-Shot Realistic and Diverse Industrial Anomaly Synthesiscited by 5
  11. 2025Foundation Visual Encoders Are Secretly Few-Shot Anomaly Detectorscited by 5
  12. 2025AnoStyler: Text-Driven Localized Anomaly Generation via Lightweight Style Transfercited by 5
  13. 2024Defect Image Sample Generation With Diffusion Prior for Steel Surface Defect Recognitioncited by 4
  14. 2026AnomalyVFM -- Transforming Vision Foundation Models into Zero-Shot Anomaly Detectorscited by 4
  15. 2025Generate Aligned Anomaly: Region-Guided Few-Shot Anomaly Image-Mask Pair Synthesis for Industrial Inspectioncited by 3
  16. 2025ISP-AD: A Large-Scale Real-World Dataset for Advancing Industrial Anomaly Detection with Synthetic and Real Defectscited by 2
  17. 2025A Comprehensive Survey for Real-World Industrial Defect Detection: Challenges, Approaches, and Prospectscited by 2
  18. 2025STAGE: Segmentation-oriented Industrial Anomaly Synthesis via Graded Diffusion with Explicit Mask Alignmentcited by 2
  19. 2026One-to-More: High-Fidelity Training-Free Anomaly Generation with Attention Controlcited by 2
  20. 2024AnomalyFactory: Regard Anomaly Generation as Unsupervised Anomaly Localizationcited by 1
  21. 2024DomainGallery: Few-shot Domain-driven Image Generation by Attribute-centric Finetuningcited by 1
  22. 2024Background-Aware Defect Generation for Robust Industrial Anomaly Detectioncited by 1
  23. 2025AnomalyHybrid: A Domain-agnostic Generative Framework for General Anomaly Detectioncited by 1
  24. 2025MAGIC: Few-Shot Mask-Guided Anomaly Inpainting with Prompt Perturbation, Spatially Adaptive Guidance, and Context Awarenesscited by 1
  25. 2025ASBench: Image Anomalies Synthesis Benchmark for Anomaly Detectioncited by 1
  26. 2025FaultDiffusion: Few-Shot Fault Time Series Generation with Diffusion Modelcited by 1
  27. 2025Anomaly Detection by Effectively Leveraging Synthetic Imagescited by 1
  28. 2026Multi-Conditioned Diffusion Synthesis of Sand Boils for Low-Resource Earthen-Levee Inspectioncited by 1
  29. 2025Bi-Grid Reconstruction for Image Anomaly Detectioncited by 0
  30. 2025CXR-AD: Component X-ray Image Dataset for Industrial Anomaly Detectioncited by 0
  31. 2025Few-Shot Anomaly-Driven Generation for Anomaly Classification and Segmentationcited by 0
  32. 2025Domain Adaptation of Attention Heads for Zero-shot Anomaly Detectioncited by 0
  33. 2025Crucial-Diff: A Unified Diffusion Model for Crucial Image and Annotation Synthesis in Data-scarce Scenarioscited by 0
  34. 2025SARD: Segmentation-Aware Anomaly Synthesis via Region-Constrained Diffusion with Discriminative Mask Guidancecited by 0
  35. 2025A Segmentation-driven Editing Method for Bolt Defect Augmentation and Detectioncited by 0
  36. 2025Training-Free Anomaly Generation via Dual-Attention Enhancement in Diffusion Modelcited by 0
  37. 2025Double Helix Diffusion for Cross-Domain Anomaly Image Generationcited by 0
  38. 2025FAST: Foreground-aware Diffusion with Accelerated Sampling Trajectory for Segmentation-oriented Anomaly Synthesiscited by 0

50 most cited of 67 All 67 (JSON)