Research map / Vision-language and multimodal / Cross-modal learning
FIND: Few-Shot Anomaly Inspection with Normal-Only Multi-Modal Data
Yiting Li, Fayao Liu, Jingyi Liao, Sichao Tian, Chuan-Sheng Foo, Xulei Yang
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 cross-modal student to the intra-modal student (ablation) | FIND · MVTec-3D-AD (10-shot) | I-AUROC: 86.3 -> 92.1 | Improved |
| add intra-modal reverse-distillation student to the cross-modal student (ablation) | FIND · MVTec-3D-AD (10-shot) | I-AUROC: 91.1 -> 92.1 | Improved |
| add cross-modal student to the intra-modal student (ablation) | FIND · MVTec-3D-AD (10-shot) | P-AUROC: 99.1 -> 99.3 | No clear effect |
| add intra-modal student to the cross-modal student (ablation) | FIND · MVTec-3D-AD (10-shot) | P-AUROC: 99.3 -> 99.3 | No clear effect |
| use surface normals instead of depth maps as 3D input (ablation) | FIND · MVTec-3D-AD (10-shot) | I-AUROC: 82.0 -> 92.1 | Improved |
| use surface normals instead of raw point clouds with PointNet (ablation) | FIND · MVTec-3D-AD (10-shot) | I-AUROC: 86.8 -> 92.1 | Improved |
| use surface normals instead of depth maps as 3D input (ablation) | FIND · MVTec-3D-AD (10-shot) | P-AUROC: 97.9 -> 99.3 | Improved |
| use FIND instead of CFM | FIND · MVTec-3D-AD (5-shot) | I-AUROC: 0.811 -> 0.899 | Improved |
| use FIND instead of CFM | FIND · MVTec-3D-AD (10-shot) | I-AUROC: 0.845 -> 0.921 | Improved |
| use FIND instead of CFM | FIND · MVTec-3D-AD (50-shot) | I-AUROC: 0.906 -> 0.952 | Improved |
| use FIND instead of CFM | FIND · MVTec-3D-AD (full-shot) | I-AUROC: 0.954 -> 0.978 | Improved |
| use FIND instead of CFM | FIND · MVTec-3D-AD (full-shot) | AUPRO@1%: 0.455 -> 0.456 | No clear effect |
Closest papers by meaning
- Beyond Single-Modal Boundary: Cross-Modal Anomaly Detection through Visual Prototype and Harmonization (CVPR 2025 · similarity 0.90)
- Rethinking Reverse Distillation for Multi-Modal Anomaly Detection (AAAI 2024 · similarity 0.90)
- Multimodal Industrial Anomaly Detection by Crossmodal Feature Mapping (CVPR 2024 · similarity 0.90)
- Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detection (CVPR 2025 · similarity 0.88)
- Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generation (CVPR 2025 · similarity 0.88)
- AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Model (AAAI 2024 · similarity 0.87)
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.
