mlsanityData to 14 Oct 2026

PEDENet: Image Anomaly Localization via Patch Embedding and Density Estimation

arXiv 2110.15525 ↗First posted 29 Oct 2021cs.CV cs.AIcited by 11

Not on the map. Why?

Abstract (arXiv)

A neural network targeting at unsupervised image anomaly localization, called the PEDENet, is proposed in this work. PEDENet contains a patch embedding (PE) network, a density estimation (DE) network, and an auxiliary network called the location prediction (LP) network. The PE network takes local image patches as input and performs dimension reduction to get low-dimensional patch embeddings via a deep encoder structure. Being inspired by the Gaussian Mixture Model (GMM), the DE network takes those patch embeddings and then predicts the cluster membership of an embedded patch. The sum of membership probabilities is used as a loss term to guide the learning process. The LP network is a Multi-layer Perception (MLP), which takes embeddings from two neighboring patches as input and predicts their relative location. The performance of the proposed PEDENet is evaluated extensively and benchmarked with that of state-of-the-art methods.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 15 of its 32 references that are in the map. First posted on arXiv 29 Oct 2021.

Builds on

15 of 32 references.

References · 15

  1. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  2. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  3. 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
  4. 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
  5. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 300
  6. 2019Deep Semi-Supervised Anomaly Detectioncited by 291
  7. 2018Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoderscited by 264
  8. 2020Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentationcited by 212
  9. 2020Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flowscited by 212
  10. 2020Explainable Deep One-Class Classificationcited by 127
  11. 2020Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detectioncited by 115
  12. 2019Towards Visually Explaining Variational Autoencoderscited by 113
3 more
  1. 2020Puzzle-AE: Novelty Detection in Images through Solving Puzzlescited by 23
  2. 2021AnomalyHop: An SSL-based Image Anomaly Localization Methodcited by 12
  3. 2021Dynamic Texture Synthesis by Incorporating Long-range Spatial and Temporal Correlationscited by 4

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 5; 2024: 4; 2025: 1.

11 papers.

  1. 2023Deep Industrial Image Anomaly Detection: A Surveycited by 148
  2. 2023Exploring Plain ViT Reconstruction for Multi-class Unsupervised Anomaly Detectioncited by 47
  3. 2024Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detectioncited by 28
  4. 2023Target before Shooting: Accurate Anomaly Detection and Localization under One Millisecond via Cascade Patch Retrievalcited by 22
  5. 2024Learning Feature Inversion for Multi-class Anomaly Detection under General-purpose COCO-AD Benchmarkcited by 16
  6. 2023UniFormaly: Towards Task-Agnostic Unified Framework for Visual Anomaly Detectioncited by 11
  7. 2023Industrial Anomaly Detection and Localization Using Weakly-Supervised Residual Transformerscited by 6
  8. 2022Self-Supervised Training with Autoencoders for Visual Anomaly Detectioncited by 3
  9. 2024Texture-AD: An Anomaly Detection Dataset and Benchmark for Real Algorithm Developmentcited by 2
  10. 2024Towards Efficient Pixel Labeling for Industrial Anomaly Detection and Localizationcited by 0
  11. 2025Towards Efficient Pixel Labeling for Industrial Anomaly Detection and Localizationcited by 0