A Recover-then-Discriminate Framework for Robust Anomaly Detection
Not on the map. Why?
Abstract (arXiv)
Anomaly detection (AD) has been extensively studied and applied in a wide range of scenarios in the recent past. However, there are still gaps between achieved and desirable levels of recognition accuracy for making AD for practical applications. In this paper, we start from an insightful analysis of two types of fundamental yet representative failure cases in the baseline model, and reveal reasons that hinder current AD methods from achieving a higher recognition accuracy. Specifically, by Case-1, we found that the main reasons detrimental to current AD methods is that the inputs to the recovery model contain a large number of detailed features to be recovered, which leads to the normal/abnormal area has-not/has been recovered into its original state. By Case-2, we surprisingly found that the abnormal area that cannot be recognized in image-level representations can be easily recognized in the feature-level representation. Based on the above observations, we propose a novel Recover-then-Discriminate (ReDi) framework for AD. ReDi takes a self-generated feature map and a selected prompted image as explicit input information to solve problems in case-1. Concurrently, a feature-level discriminative network is proposed to enhance abnormal differences between the recovered representation and the input representation. Extensive experimental results on two popular yet challenging AD datasets validate that ReDi achieves the new state-of-the-art accuracy.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 46 of its 73 references that are in the map. First posted on arXiv 7 Jun 2024.
Builds on
46 of 73 references.
References · 46
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2016Wide Residual Networkscited by 3,057
- 2021Learning to Prompt for Vision-Language Modelscited by 2,269
- 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
- 2015Variational Inference with Normalizing Flowscited by 2,161
- 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,419
- 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
- 2021Resolution-robust Large Mask Inpainting with Fourier Convolutionscited by 730
- 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 591
- 2021Masked Feature Prediction for Self-Supervised Visual Pre-Trainingcited by 573
34 more
- 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 560
- 2021DenseCLIP: Language-Guided Dense Prediction with Context-Aware Promptingcited by 530
- 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 529
- 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 510
- 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 419
- 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
- 2021How Much Can CLIP Benefit Vision-and-Language Tasks?cited by 321
- 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 320
- 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 297
- 2020Multiresolution Knowledge Distillation for Anomaly Detectioncited by 285
- 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 270
- 2022A Unified Model for Multi-class Anomaly Detectioncited by 267
- 2018Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoderscited by 263
- 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
- 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 225
- 2022MAT: Mask-Aware Transformer for Large Hole Image Inpaintingcited by 216
- 2020Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flowscited by 212
- 2021Fully Convolutional Cross-Scale-Flows for Image-based Defect Detectioncited by 117
- 2024RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detectioncited by 116
- 2020Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detectioncited by 115
- 2021Mixed supervision for surface-defect detection: from weakly to fully supervised learningcited by 101
- 2021Divide-and-Assemble: Learning Block-wise Memory for Unsupervised Anomaly Detectioncited by 83
- 2023PyramidFlow: High-Resolution Defect Contrastive Localization using Pyramid Normalizing Flowcited by 74
- 2022Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detectioncited by 72
- 2021CTNet: Context-based Tandem Network for Semantic Segmentationcited by 43
- 2021SQUID: Deep Feature In-Painting for Unsupervised Anomaly Detectioncited by 34
- 2021Semi-orthogonal Embedding for Efficient Unsupervised Anomaly Segmentationcited by 26
- 2020Puzzle-AE: Novelty Detection in Images through Solving Puzzlescited by 23
- 2023Context Disentangling and Prototype Inheriting for Robust Visual Groundingcited by 23
- 2021AnoSeg: Anomaly Segmentation Network Using Self-Supervised Learningcited by 21
- 2021Reconstruction Student with Attention for Student-Teacher Pyramid Matchingcited by 20
- 2022Visual Anomaly Detection Via Partition Memory Bank Module and Error Estimationcited by 20
- 2021Self-Regulation for Semantic Segmentationcited by 18
- 2022ADPS: Asymmetric Distillation Post-Segmentation for Image Anomaly Detectioncited by 5
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.