AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Semantic Novelty Detection via Relational Reasoning

Francesco Cappio Borlino, Silvia Bucci, Tatiana Tommasi

ECCV 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Industrial anomaly detection · 3 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
use learned transformer relational module instead of Max aggregation (ablation)ReSeND · PACS multi-sourceavg AUROC: 0.726 -> 0.765Improved
use learned transformer relational module instead of Sum aggregation (ablation)ReSeND · PACS multi-sourceavg AUROC: 0.530 -> 0.765Improved
use learned transformer relational module instead of Concat aggregation (ablation)ReSeND · PACS multi-sourceavg AUROC: 0.736 -> 0.765Improved
train with MSE regression instead of cross-entropy classification (ablation)ReSeND · Open-set DG benchmarksAUROCImproved
add ReSeND score to SagNet MSPSagNet · Open-set DG single-sourceAUROC: 0.643 -> 0.700Improved
add ReSeND score to Diversify MSPDiversify · Open-set DG single-sourceAUROC: 0.643 -> 0.691Improved
add ReSeND score to DAML MSPDAML · Open-set DG multi-sourceAUROC: 0.657 -> 0.722Improved
add ReSeND score to SWAD MSPSWAD · Open-set DG multi-sourceAUROC: 0.570 -> 0.700Improved
add ReSeND score to SWAD MSPSWAD · Open-set DG multi-sourceH-score: 42.85 -> 57.05Improved
add ReSeND score to DAML MSPDAML · Open-set DG multi-sourceH-score: 52.99 -> 57.93Improved
finetune Mahalanobis on the support setMahalanobis · PACS multi-sourceavg AUROC: 0.616 -> 0.665Improved
finetune Gram on the support setGram · PACS multi-sourceavg AUROC: 0.656 -> 0.677Improved

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.

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

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