Research map / Label-efficient and robust learning / OOD and anomaly detection
FastRecon: Few-shot Industrial Anomaly Detection via Fast Feature Reconstruction
Zheng Fang, Xiaoyang Wang, Haocheng Li, Jiejie Liu, Qiugui Hu, Jimin Xiao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace ridge regression penalty with distribution regularization (ablation) | FastRecon · MVTec AD (2-shot) | image-level AUROC: 87.16 -> 90.97 | Improved |
| replace ridge regression penalty with distribution regularization (ablation) | FastRecon · MVTec AD (8-shot) | pixel-level AUROC: 95.42 -> 97.27 | Improved |
| replace ridge regression penalty with distribution regularization (ablation) | FastRecon · MVTec AD (4-shot) | image-level AUROC: 89.55 -> 94.24 | Improved |
| set regularization coefficient lambda to 0.3 (ablation) | FastRecon · MVTec AD (4-/8-shot) | AUROC | Improved |
| use input resolution 368 x 368 (ablation) | FastRecon · MVTec AD | AUROC | Improved |
| vary coreset sampling rate s (ablation) | FastRecon · MVTec AD | AUROC | Mixed |
| replace aggregated-feature training with closed-form regression (no training on aggregated features) | FastRecon vs RegAD · MVTec AD (k=2) | training time (s): 660.80 -> 0.52 | Improved |
Later papers that compared with this method
1 later accepted paper reports FastRecon as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detection (CVPR 2024 · better on 0 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FastRef: Fast Prototype Refinement for Few-shot Industrial Anomaly Detection (CVPR 2026 · similarity 0.93)
- RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection (CVPR 2024 · similarity 0.88)
- AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Model (AAAI 2024 · similarity 0.87)
- One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection (NeurIPS 2024 · similarity 0.87)
- 3CAD: A Large-Scale Real-World 3C Product Dataset for Unsupervised Anomaly Detection (AAAI 2025 · similarity 0.86)
- Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generation (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
