AI Paper Atlas

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

SANFlow: Semantic-Aware Normalizing Flow for Anomaly Detection

Daehyun Kim, Sungyong Baik, Tae Hyun Kim

NeurIPS 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection · 3 citations (OpenAlex)

Read the paper

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
train with all three losses (add L_BCE and L_KL) vs plain NLL (ablation)NF baseline · STC image-wiseAUROC: 72.6 -> 76.1Improved
train with all three losses vs plain NLL (ablation)NF baseline · MVTec image-wiseAUROC: 98.3 -> 98.7Improved
add L_KL on top of Model (1d) (ablation)SANFlow · STC image-wiseAUROC: 74.2 -> 76.1Improved
add L_BCE on top of Model (1c) (ablation)SANFlow · STC image-wiseAUROC: 73.1 -> 76.1Improved
add L_KL only (Model 1c) to Model (1b) (ablation)NF with NLL Eq.3&5 · MVTec image-wiseAUROC: 98.2 -> 98.3No clear effect
add synthetic anomaly augmentation (ablation)SANFlow · STC image-wiseAUROC: 73.6 -> 74.2Improved
add synthetic anomaly augmentation (ablation)SANFlow · MVTec image-wiseAUROC: 98.4 -> 98.5Improved
estimate variance with network, mean fixed (ablation)SANFlow · STC image-wiseAUROC: 75.9 -> 76.1Improved
estimate variance with network, mean fixed (ablation)SANFlow · MVTec image-wiseAUROC: 97.6 -> 98.7Improved
also estimate mean (ablation)SANFlow · STC image-wiseAUROC: 76.1 -> 74.5Hurt
mean margin 1.0 vs 0.5 vs 1.5 (ablation)SANFlow · MVTec image-wiseAUROC: 98.6, 98.7, 98.1Mixed
add statistics prediction (ablation)CFLOW-AD · STC image-wiseAUROC: 72.63 -> 73.6Improved

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