Research map / Label-efficient and robust learning / OOD and anomaly detection
Dual-Prototype-Guided Multi-task Learning for Unsupervised Anomaly Detection and Classification
Qianhao Luo, Jiajia Mi, Mingtao Yan, JingSheng Liu, ShuYang Pang, Weiling Li
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 |
|---|---|---|---|
| add Dual-Prototype Representation Module (DPRM) to baseline (ablation) | PG-SFD baseline · MVTec-AD | P-AUPRO: 20.5 -> 94.5 | Improved |
| add DPRM to baseline (ablation) | PG-SFD baseline · MVTec-AD | Cls-F1: 40.5 -> 53.0 | Improved |
| add Differential Gated Interaction (DGI) (ablation) | V1 (DPRM) · MVTec-AD | Cls-F1: 53.0 -> 57.3 | Improved |
| add DGI (ablation) | V1 (DPRM) · MVTec-AD | I-AUROC / P-AUPRO: 94.0 -> 93.8; 94.5 -> 94.1 | No clear effect |
| add Geometry-Regularized Optimization (GRO) (ablation) | V2 (DPRM + DGI) · MVTec-AD | I-AUROC: 93.8 -> 99.4 | Improved |
| add GRO (ablation) | V2 (DPRM + DGI) · MVTec-AD | Cls-F1: 57.3 -> 61.6 | Improved |
| joint end-to-end training instead of two-stage locate-then-classify | PG-SFD · MVTec-AD | Cls-F1: 48.5 -> 61.6 | Improved |
| PG-SFD versus DinoCLS on classification F1 | PG-SFD · MVTec-AD | Cls-F1: 60.9 -> 61.6 | Improved |
| PG-SFD versus DinoCLS on medical and VisA classification | PG-SFD · Uni-Medical, VisA | Cls-F1: Uni-Medical 77.6 vs 80.84; VisA 79.1 vs 80.44 | Hurt |
| normal prototype count K = 6 (ablation) | PG-SFD · MVTec-AD | balanced performance | Improved |
| PG-SFD versus INPformer on pixel AUPRO | PG-SFD · MVTec-AD | AUPRO: 96.0 vs 95.6 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 on most shared results.
- Dinomaly — Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection (CVPR 2025) · better on 1 of 3 shared results
- SimpleNet — SimpleNet: A Simple Network for Image Anomaly Detection and Localization (CVPR 2023) · 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
- AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios (CVPR 2025 · similarity 0.87)
- Beyond Normal References: Discriminative Few-Shot Anomaly Detection (NeurIPS 2026 · similarity 0.86)
- UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection (CVPR 2025 · similarity 0.86)
- AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models (AAAI 2024 · similarity 0.86)
- ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining (NeurIPS 2025 · similarity 0.85)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.85)
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.
