AI Paper Atlas

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

Utilizing Patch-level Category Activation Patterns for Multiple Class Novelty Detection

Poojan Oza, Vishal M. Patel

ECCV 2020 · Label-efficient and robust learning › OOD and anomaly detection · approach: Deep one-class and tabular anomaly detection · 7 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
add patch-level mismatch novelty network on top of global network (ablation)VGG16 global network · Overall (4 datasets, VGG16)AUC: 0.841 -> 0.883Improved
add global network to local-only network (ablation)BagNet local network · Overall (4 datasets, VGG16)AUC: 0.739 -> 0.883Improved
replace naive Global+Local score fusion with mismatch-trained novelty network (ablation)Global+Local · Overall (4 datasets, VGG16)AUC: 0.837 -> 0.883Improved
naive fusion of global and local scores (no mismatch training) (ablation)Global Only · Overall (4 datasets, VGG16)AUC: 0.841 -> 0.837No clear effect
add mismatch-trained novelty network (ablation)Global Only VGG16 · Stanford DogsAUC: 0.766 -> 0.827Improved
add reference dataset regularization on penultimate activationsProposed · Overall (4 datasets, VGG16 and AlexNet avg)AUC: 0.857 -> 0.879Improved
add reference dataset regularizationProposed · Stanford Dogs VGG16AUC: 0.827 -> 0.873Improved
add reference dataset regularizationProposed · FounderType-200 AlexNetAUC: 0.798 -> 0.801No clear effect
use proposed method without reference data instead of DTMND without reference dataDTMND · Overall (4 datasets)AUC: 0.80 -> 0.857Improved
use reference dataset via simple fine-tuningFine-tune · Overall (4 datasets)AUC: 0.801 -> 0.800No clear effect
replace Fine-tune score with OpenMaxFine-tune · Overall (4 datasets)AUC: 0.801 -> 0.809Improved
use proposed method versus Fine-tune baselineFine-tune · FounderType-200 VGG16AUC: 0.841 -> 0.876Mixed

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