Research map / Label-efficient and robust learning / OOD and anomaly detection
Wave-MambaAD: Wavelet-driven State Space Model for Multi-class Unsupervised Anomaly Detection
Qiao Zhang, Mingwen Shao, Xinyuan Chen, Xiang Lv, Kai Xu
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 Wavelet-Mamba module instead of HSS (ablation) | Wave-MambaAD · MVTec-AD | mAD: 84.2 -> 85.6 | Improved |
| add high-frequency state space model (ablation) | Wave-MambaAD · MVTec-AD | mAD: 84.2 -> 85.6 | Improved |
| add low-frequency state space model (ablation) | Wave-MambaAD · MVTec-AD | mAD: 84.6 -> 85.6 | Improved |
| use DSE block instead of MLP (ablation) | Wave-MambaAD · MVTec-AD | mAD: 84.9 -> 85.6 | Improved |
| use horizontal, vertical and diagonal scanning rather than horizontal only (ablation) | Wave-MambaAD · MVTec-AD | mAD: 85.0 -> 85.6 | Improved |
| use directional scanning rather than vertical only (ablation) | Wave-MambaAD · MVTec-AD | mAD: 85.3 -> 85.6 | Improved |
| add diagonal scan to HFSS (ablation) | Wave-MambaAD · MVTec-AD | mAD: 85.5 -> 85.6 | Improved |
| channel-wise scanning in LFSS instead of spatial-wise (ablation) | Wave-MambaAD · MVTec-AD | mAD: 84.9 -> 85.6 | Improved |
| Wave-MambaAD over MambaAD | Wave-MambaAD · MVTec-AD (multi-class) | mAD: 85.4 -> 85.6 | Improved |
| Wave-MambaAD over MambaAD | Wave-MambaAD · VisA (multi-class) | mAD: 78.3 -> 79.0 | Improved |
| Wave-MambaAD over MambaAD | Wave-MambaAD · MPDD (multi-class) | mAD: 76.3 -> 79.1 | Improved |
| Wave-MambaAD over MambaAD efficiency | Wave-MambaAD · Efficiency | Params (M): 25.7 -> 22.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- RealNet — RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection (CVPR 2024) · better on 3 of 3 shared results
- SimpleNet — SimpleNet: A Simple Network for Image Anomaly Detection and Localization (CVPR 2023) · better on 3 of 3 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
- MambaAD: Exploring State Space Models for Multi-class Unsupervised Anomaly Detection (NeurIPS 2024 · similarity 0.91)
- A Unified Latent Schrodinger Bridge Diffusion Model for Unsupervised Anomaly Detection and Localization (CVPR 2025 · similarity 0.85)
- Hyperbolic Anomaly Detection (CVPR 2024 · similarity 0.85)
- A Diffusion-Based Framework for Multi-Class Anomaly Detection (AAAI 2024 · similarity 0.85)
- ScatterAD: Temporal-Topological Scattering Mechanism for Time Series Anomaly Detection (NeurIPS 2025 · similarity 0.84)
- DZAD: Diffusion-based Zero-shot Anomaly Detection (AAAI 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.
