mlsanityData to 14 Oct 2026

Clear Memory-Augmented Auto-Encoder for Surface Defect Detection

arXiv 2208.03879 ↗First posted 8 Aug 2022cs.CV cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

In surface defect detection, due to the extreme imbalance in the number of positive and negative samples, positive-samples-based anomaly detection methods have received more and more attention. Specifically, reconstruction-based methods are the most popular. However, existing methods are either difficult to repair abnormal foregrounds or reconstruct clear backgrounds. Therefore, we propose a clear memory-augmented auto-encoder (CMA-AE). At first, we propose a novel clear memory-augmented module (CMAM), which combines the encoding and memoryencoding in a way of forgetting and inputting, thereby repairing abnormal foregrounds and preserving clear backgrounds. Secondly, a general artificial anomaly generation algorithm (GAAGA) is proposed to simulate anomalies that are as realistic and feature-rich as possible. At last, we propose a novel multi scale feature residual detection method (MSFR) for defect segmentation, which makes the defect location more accurate. Extensive comparison experiments demonstrate that CMA-AE achieves state-of-the-art detection accuracy and shows great potential in industrial applications.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 21 of its 48 references that are in the map. First posted on arXiv 8 Aug 2022.

Builds on

21 of 48 references.

References · 21

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  4. 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,436
  5. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  6. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 564
  7. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  8. 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
  9. 2015Adversarial Autoencoderscited by 364
  10. 2014Memory Networkscited by 363
  11. 2018Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoderscited by 264
  12. 2019OCGAN: One-class Novelty Detection Using GANs with Constrained Latent Representationscited by 199
9 more
  1. 2019Segmentation-Based Deep-Learning Approach for Surface-Defect Detectioncited by 145
  2. 2019Skip-GANomaly: Skip Connected and Adversarially Trained Encoder-Decoder Anomaly Detectioncited by 100
  3. 2018Generative Probabilistic Novelty Detection with Adversarial Autoencoderscited by 99
  4. 2021Divide-and-Assemble: Learning Block-wise Memory for Unsupervised Anomaly Detectioncited by 83
  5. 2020Iterative energy-based projection on a normal data manifold for anomaly localizationcited by 47
  6. 2022InsetGAN for Full-Body Image Generationcited by 45
  7. 2016Scaling Memory-Augmented Neural Networks with Sparse Reads and Writescited by 28
  8. 2020Unsupervised Video Anomaly Detection via Normalizing Flows with Implicit Latent Featurescited by 23
  9. 2020TrustMAE: A Noise-Resilient Defect Classification Framework using Memory-Augmented Auto-Encoders with Trust Regionscited by 21

Built on by

Cited by year

Citing papers per year: 2022: 1.

1 paper.

  1. 2022Normal Reference Attention and Defective Feature Perception Network for Surface Defect Detectioncited by 9