Research map / Label-efficient and robust learning / Recognition and long-tail
IMMoE: Incomplete Multi-View Anomaly Detection via Mixture of View Experts Fusion
Lei Hu
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 Local Anomaly Enhancement Encoder (feature dropout) (ablation) | IMMoE · RIMAD | Image-AUROC: 76.9 -> 81.3 | Improved |
| add Multi-view Expert Fusion (ablation) | IMMoE · RIMAD | Image-AUROC: 81.3 -> 84.0 | Improved |
| add Area Adaptive Loss (ablation) | IMMoE · RIMAD | Image-AUROC: 84.0 -> 84.7 | Improved |
| add Area Adaptive Loss (ablation) | IMMoE · RIMAD | Pixel-AP: 32.3 -> 33.1 | Improved |
| add feature dropout rate 0.1 vs none (ablation) | IMMoE · RIMAD | Image-AUROC: 83.0 -> 84.7 | Improved |
| dropout rate too high (0.3) (ablation) | IMMoE · RIMAD | Image-AUROC: 84.7 -> 76.9 | Hurt |
| dropout rate 0.2 vs 0.1 (ablation) | IMMoE · RIMAD | Image-AUROC: 84.7 -> 84.2 | No clear effect |
| more input views (1 to 5) (ablation) | IMMoE · RIMAD | Image-AUROC: 80.3 -> 84.7 | Improved |
| increase mask ratio from 25% to 75% (ablation) | IMMoE · RIMAD | Image-AUROC: 77.9 -> 87.3 | Improved |
| IMMoE vs INP-Former on incomplete views | IMMoE · RIMAD | Pixel-AP: 29.6 -> 33.1 | Improved |
| IMMoE vs MVAD (multi-view baseline) | IMMoE · RIMAD | Image-AUROC: 79.3 -> 84.7 | Improved |
| IMMoE on complete views | IMMoE · Real-IAD | Image-AUROC: 89.3 -> 89.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Dinomaly — Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection (CVPR 2025) · better on 4 of 4 shared results
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
- Unveiling Multi-View Anomaly Detection: Intra-view Decoupling and Inter-view Fusion (AAAI 2025 · similarity 0.90)
- Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detection (CVPR 2024 · similarity 0.86)
- AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models (AAAI 2024 · similarity 0.86)
- Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detection (CVPR 2025 · similarity 0.85)
- UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection (CVPR 2025 · similarity 0.84)
- Towards Training-free Anomaly Detection with Vision and Language Foundation Models (CVPR 2025 · similarity 0.84)
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.
