AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Hierarchical Vector Quantized Transformer for Multi-class Unsupervised Anomaly Detection

Ruiying Lu, YuJie Wu, Long Tian, Dongsheng Wang, Bo Chen, Xiyang Liu, Ruimin Hu

NeurIPS 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection · 22 citations (OpenAlex)

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 changedWhereMeasuredEffect
replace continuous latent features with vector-quantized normal prototypes (ablation)HVQ-Trans on MVTec-AD · MVTec-ADimage-level detection AUROC: 70.5 to 98.0Improved
use hierarchical VQLayers rather than a single VQLayer (ablation)HVQ-Trans on MVTec-AD · MVTec-ADimage-level detection AUROC: 96.2 to 97.1Improved
add codebook switching (ablation)HVQ-Trans with hierarchical VQLayers · MVTec-ADimage-level detection AUROC: 97.1 to 97.2Improved
add POT-based scoring (ablation)HVQ-Trans with hierarchical VQLayers · MVTec-ADimage-level detection AUROC: 97.1 to 97.4Improved
add codebook and expert switching (ablation)HVQ-Trans with hierarchical VQLayers · MVTec-ADimage-level detection AUROC: 97.1 to 97.6Improved
add POT to the model with codebook and expert switching (ablation)HVQ-Trans · MVTec-ADimage-level detection AUROC: 97.6 to 98.0Improved
concatenate global prototypes with earlier-layer features for quantization (ablation)HVQ-Trans hierarchy · MVTec-ADimage-level detection AUROC: 97.2 to 98.0Improved
use 512 prototypes rather than 1024 or 256 (ablation)HVQ-Trans · MVTec-ADimage-level detection AUROC: 1024: 97.1; 512: 98.0; 256: 97.0Mixed
use HVQ-Trans instead of UniADone-for-all anomaly detection · MVTec-AD, mean image-levelAUROC: 96.5 to 98.0Improved
use HVQ-Trans instead of UniADone-for-all anomaly localization · MVTec-AD, mean pixel-wiseAUROC: 96.8 to 97.3Improved
use HVQ-Trans instead of UniADone-for-all anomaly detection · VisA, mean image-levelAUROC: 91.9 to 93.2Improved
use HVQ-Trans instead of UniADone-for-all anomaly localization · VisA, mean pixel-wiseAUROC: 98.6 to 98.7Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy