mlsanityData to 7 Oct 2026

ADPS: Asymmetric Distillation Post-Segmentation for Image Anomaly Detection

arXiv 2210.10495 ↗First posted 19 Oct 2022cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

Knowledge Distillation-based Anomaly Detection (KDAD) methods rely on the teacher-student paradigm to detect and segment anomalous regions by contrasting the unique features extracted by both networks. However, existing KDAD methods suffer from two main limitations: 1) the student network can effortlessly replicate the teacher network's representations, and 2) the features of the teacher network serve solely as a ``reference standard" and are not fully leveraged. Toward this end, we depart from the established paradigm and instead propose an innovative approach called Asymmetric Distillation Post-Segmentation (ADPS). Our ADPS employs an asymmetric distillation paradigm that takes distinct forms of the same image as the input of the teacher-student networks, driving the student network to learn discriminating representations for anomalous regions. Meanwhile, a customized Weight Mask Block (WMB) is proposed to generate a coarse anomaly localization mask that transfers the distilled knowledge acquired from the asymmetric paradigm to the teacher network. Equipped with WMB, the proposed Post-Segmentation Module (PSM) is able to effectively detect and segment abnormal regions with fine structures and clear boundaries. Experimental results demonstrate that the proposed ADPS outperforms the state-of-the-art methods in detecting and segmenting anomalies. Surprisingly, ADPS significantly improves Average Precision (AP) metric by 9% and 20% on the MVTec AD and KolektorSDD2 datasets, respectively.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 45 of its 79 references that are in the map. First posted on arXiv 19 Oct 2022.

Builds on

45 of 79 references.

References · 45

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2016Wide Residual Networkscited by 3,057
  4. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
  5. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 591
  6. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  7. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 560
  8. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 529
  9. 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 510
  10. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 419
  11. 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
  12. 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
33 more
  1. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 297
  2. 2020Multiresolution Knowledge Distillation for Anomaly Detectioncited by 285
  3. 2020Learning Memory-guided Normality for Anomaly Detectioncited by 278
  4. 2022A Unified Model for Multi-class Anomaly Detectioncited by 267
  5. 2018Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoderscited by 263
  6. 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 225
  7. 2020Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentationcited by 212
  8. 2020Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flowscited by 212
  9. 2021A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challengescited by 154
  10. 2018Deep Autoencoding Models for Unsupervised Anomaly Segmentation in Brain MR Imagescited by 146
  11. 2019Segmentation-Based Deep-Learning Approach for Surface-Defect Detectioncited by 145
  12. 2021Student-Teacher Feature Pyramid Matching for Anomaly Detectioncited by 145
  13. 2018Efficient GAN-Based Anomaly Detectioncited by 122
  14. 2021Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detectioncited by 119
  15. 2021Fully Convolutional Cross-Scale-Flows for Image-based Defect Detectioncited by 117
  16. 2020Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detectioncited by 115
  17. 2022Catching Both Gray and Black Swans: Open-set Supervised Anomaly Detectioncited by 104
  18. 2021Mixed supervision for surface-defect detection: from weakly to fully supervised learningcited by 101
  19. 2022Anomaly Detection in Autonomous Driving: A Surveycited by 87
  20. 2021Divide-and-Assemble: Learning Block-wise Memory for Unsupervised Anomaly Detectioncited by 83
  21. 2018Anomaly Detection using One-Class Neural Networkscited by 82
  22. 2022Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detectioncited by 72
  23. 2022Deep Learning for Unsupervised Anomaly Localization in Industrial Images: A Surveycited by 67
  24. 2021Feature Encoding with AutoEncoders for Weakly-supervised Anomaly Detectioncited by 42
  25. 2021SQUID: Deep Feature In-Painting for Unsupervised Anomaly Detectioncited by 34
  26. 2022Switchable Online Knowledge Distillationcited by 29
  27. 2021Semi-orthogonal Embedding for Efficient Unsupervised Anomaly Segmentationcited by 26
  28. 2020Puzzle-AE: Novelty Detection in Images through Solving Puzzlescited by 23
  29. 2021AnoSeg: Anomaly Segmentation Network Using Self-Supervised Learningcited by 21
  30. 2022Visual Anomaly Detection Via Partition Memory Bank Module and Error Estimationcited by 20
  31. 2022SiRi: A Simple Selective Retraining Mechanism for Transformer-based Visual Groundingcited by 12
  32. 2015PCA leverage: outlier detection for high-dimensional functional magnetic resonance imaging datacited by 3
  33. 2022Self-Supervised Guided Segmentation Framework for Unsupervised Anomaly Detectioncited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 2; 2026: 1.

5 papers.

  1. 2024Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detectioncited by 16
  2. 2025Dual Learning with Dynamic Knowledge Distillation and Soft Alignment for Partially Relevant Video Retrievalcited by 3
  3. 2024A Recover-then-Discriminate Framework for Robust Anomaly Detectioncited by 2
  4. 2025A Comprehensive Survey for Real-World Industrial Defect Detection: Challenges, Approaches, and Prospectscited by 2
  5. 2026URA-Net: Uncertainty-Integrated Anomaly Perception and Restoration Attention Network for Unsupervised Anomaly Detectioncited by 1