mlsanityData to 7 Oct 2026

What makes a good data augmentation for few-shot unsupervised image anomaly detection?

arXiv 2304.03294 ↗First posted 6 Apr 2023cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Data augmentation is a promising technique for unsupervised anomaly detection in industrial applications, where the availability of positive samples is often limited due to factors such as commercial competition and sample collection difficulties. In this paper, how to effectively select and apply data augmentation methods for unsupervised anomaly detection is studied. The impact of various data augmentation methods on different anomaly detection algorithms is systematically investigated through experiments. The experimental results show that the performance of different industrial image anomaly detection (termed as IAD) algorithms is not significantly affected by the specific data augmentation method employed and that combining multiple data augmentation methods does not necessarily yield further improvements in the accuracy of anomaly detection, although it can achieve excellent results on specific methods. These findings provide useful guidance on selecting appropriate data augmentation methods for different requirements in IAD.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 28 of its 38 references that are in the map. First posted on arXiv 6 Apr 2023.

Builds on

28 of 38 references.

References · 28

  1. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  2. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
  3. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 591
  4. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 529
  5. 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 510
  6. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 419
  7. 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
  8. 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 320
  9. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 297
  10. 2020Multiresolution Knowledge Distillation for Anomaly Detectioncited by 285
  11. 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 225
  12. 2021VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localizationcited by 222
16 more
  1. 2020Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentationcited by 212
  2. 2020Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flowscited by 212
  3. 2021Student-Teacher Feature Pyramid Matching for Anomaly Detectioncited by 145
  4. 2022CFA: Coupled-hypersphere-based Feature Adaptation for Target-Oriented Anomaly Localizationcited by 135
  5. 2020Learning and Evaluating Representations for Deep One-class Classificationcited by 133
  6. 2022Registration based Few-Shot Anomaly Detectioncited by 133
  7. 2021Fully Convolutional Cross-Scale-Flows for Image-based Defect Detectioncited by 117
  8. 2022Image Data Augmentation for Deep Learning: A Surveycited by 96
  9. 2022DSR -- A dual subspace re-projection network for surface anomaly detectioncited by 89
  10. 2021A Hierarchical Transformation-Discriminating Generative Model for Few Shot Anomaly Detectioncited by 60
  11. 2020DFR: Deep Feature Reconstruction for Unsupervised Anomaly Segmentationcited by 53
  12. 2022MAEDAY: MAE for few and zero shot AnomalY-Detectioncited by 30
  13. 2021Semi-orthogonal Embedding for Efficient Unsupervised Anomaly Segmentationcited by 26
  14. 2020Anomaly localization by modeling perceptual featurescited by 21
  15. 2021Reconstruction Student with Attention for Student-Teacher Pyramid Matchingcited by 20
  16. 2021Anomaly Detection via Self-organizing Mapcited by 18

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 0; 2025: 3; 2026: 2.

8 papers.

  1. 2023Real3D-AD: A Dataset of Point Cloud Anomaly Detectioncited by 87
  2. 2023Visual and Textual Prior Guided Mask Assemble for Few-Shot Segmentation and Beyondcited by 19
  3. 2025Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detectioncited by 12
  4. 2023Optimizing PatchCore for Few/many-shot Anomaly Detectioncited by 10
  5. 2025MAGIC: Few-Shot Mask-Guided Anomaly Inpainting with Prompt Perturbation, Spatially Adaptive Guidance, and Context Awarenesscited by 1
  6. 2025FracAug: Fractional Augmentation boost Graph-level Anomaly Detection under Limited Supervisioncited by 0
  7. 2026ConceptADapt: Concept-guided Adaptive Feature Reconstruction with Dynamic Attention for Few-Shot Industrial Anomaly Detectioncited by 0
  8. 2026From Benchmark Performance to Tool Deployment: Human-in-the-Loop Anomaly Detectioncited by 0