Research map / Label-efficient and robust learning / OOD and anomaly detection
Commonality in Few: Few-Shot Multimodal Anomaly Detection via Hypergraph-Enhanced Memory
Yuxuan Lin, Hanjing Yan, Xuan Tong, Yang Chang, Huanzhen Wang, Ziheng Zhou, Shuyong Gao, Yan Wang, Wenqiang Zhang
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 structure-guided memory sampling (SGMS) (ablation) | CIF base · MVTec 3D-AD 1-shot | I-AUROC: 68.6 -> 71.2 | Improved |
| add SGMS (ablation) | CIF base · MVTec 3D-AD 1-shot | AUPRO: 76.1 -> 85.0 | Improved |
| add hyperedge-guided memory search (HGMS) (ablation) | CIF with SGMS · MVTec 3D-AD 1-shot | I-AUROC: 71.2 -> 71.7 | Improved |
| add Bi-TF-MP (ablation) | CIF with SGMS+HGMS · MVTec 3D-AD 1-shot | I-AUROC: 71.7 -> 72.0 | Improved |
| add Bi-TF-MP (ablation) | CIF with SGMS+HGMS · MVTec 3D-AD 1-shot | AUPRO: 86.0 -> 86.1 | No clear effect |
| use 4 hyperedges vs 8 in SAHC (ablation) | CIF · MVTec 3D-AD 1-shot | I-AUROC / AUPRO | Mixed |
| message passing layers beyond L=1 at alpha 0.9 (ablation) | Bi-TF-MP · MVTec 3D-AD 1-shot | I-AUROC | No clear effect |
| CIF vs training-free Patchcore+FPFH | CIF · MVTec 3D-AD 1-shot | I-AUROC: 59.9 -> 72.0 | Improved |
| CIF vs training-based methods on MVTec 3D-AD | CIF · MVTec 3D-AD 1-shot | I-AUROC: 73.9 -> 72.0 | Hurt |
| CIF vs best baseline on localization | CIF · Eyecandies 1-shot | AUPRO: 79.9 -> 69.2 | Hurt |
| CIF vs M3DM/CFM on detection | CIF · Eyecandies 1-shot | I-AUROC: 64.1 -> 69.5 | Improved |
Closest papers by meaning
- Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection (AAAI 2024 · similarity 0.87)
- Multimodal Industrial Anomaly Detection by Crossmodal Feature Mapping (CVPR 2024 · similarity 0.87)
- AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models (AAAI 2024 · similarity 0.86)
- MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled Images (ICLR 2024 · similarity 0.86)
- UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection (CVPR 2025 · similarity 0.86)
- Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detection (CVPR 2025 · 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.
