mlsanityData to 14 Oct 2026

LADMIM: Logical Anomaly Detection with Masked Image Modeling in Discrete Latent Space

arXiv 2410.10234 ↗First posted 14 Oct 2024cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Detecting anomalies such as an incorrect combination of objects or deviations in their positions is a challenging problem in unsupervised anomaly detection (AD). Since conventional AD methods mainly focus on local patterns of normal images, they struggle with detecting logical anomalies that appear in the global patterns. To effectively detect these challenging logical anomalies, we introduce Logical Anomaly Detection with Masked Image Modeling (LADMIM), a novel unsupervised AD framework that harnesses the power of masked image modeling and discrete representation learning. Our core insight is that predicting the missing region forces the model to learn the long-range dependencies between patches. Specifically, we formulate AD as a mask completion task, which predicts the distribution of discrete latents in the masked region. As a distribution of discrete latents is invariant to the low-level variance in the pixel space, the model can desirably focus on the logical dependencies in the image, which improves accuracy in the logical AD. We evaluate the AD performance on five benchmarks and show that our approach achieves compatible performance without any pre-trained segmentation models. We also conduct comprehensive experiments to reveal the key factors that influence logical AD performance.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 36 of its 49 references that are in the map. First posted on arXiv 14 Oct 2024.

Builds on

36 of 49 references.

References · 36

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  4. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  5. 2017Neural Discrete Representation Learningcited by 4,712
  6. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,375
  7. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  8. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
  9. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 1,039
  10. 2022Autoregressive Image Generation using Residual Quantizationcited by 722
  11. 2020Jukebox: A Generative Model for Musiccited by 542
  12. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
24 more
  1. 2022BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizerscited by 303
  2. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 300
  3. 2022Context Autoencoder for Self-Supervised Representation Learningcited by 297
  4. 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 227
  5. 2023EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latenciescited by 160
  6. 2023Deep Industrial Image Anomaly Detection: A Surveycited by 148
  7. 2022Learning Vector-Quantized Item Representation for Transferable Sequential Recommenderscited by 112
  8. 2023BMAD: Benchmarks for Medical Anomaly Detectioncited by 82
  9. 2024A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localizationcited by 74
  10. 2023Hierarchical Vector Quantized Transformer for Multi-class Unsupervised Anomaly Detectioncited by 66
  11. 2023Towards Accurate Image Coding: Improved Autoregressive Image Generation with Dynamic Vector Quantizationcited by 60
  12. 2022Self-Supervised Masking for Unsupervised Anomaly Detection and Localizationcited by 43
  13. 2022MAEDAY: MAE for few and zero shot AnomalY-Detectioncited by 30
  14. 2023Not All Image Regions Matter: Masked Vector Quantization for Autoregressive Image Generationcited by 28
  15. 2023Component-aware anomaly detection framework for adjustable and logical industrial visual inspectioncited by 22
  16. 2023Learning Global-Local Correspondence with Semantic Bottleneck for Logical Anomaly Detectioncited by 21
  17. 2023Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detectioncited by 19
  18. 2022Are we certain it's anomalous?cited by 13
  19. 2024CSAD: Unsupervised Component Segmentation for Logical Anomaly Detectioncited by 12
  20. 2022Adaptive Discrete Communication Bottlenecks with Dynamic Vector Quantizationcited by 11
  21. 2023SLSG: Industrial Image Anomaly Detection by Learning Better Feature Embeddings and One-Class Classificationcited by 10
  22. 2022Siamese Transition Masked Autoencoders as Uniform Unsupervised Visual Anomaly Detectorcited by 9
  23. 2024PUAD: Frustratingly Simple Method for Robust Anomaly Detectioncited by 7
  24. 2022Hierarchical Residual Learning Based Vector Quantized Variational Autoencoder for Image Reconstruction and Generationcited by 6

Built on by

Not cited yet.