AI Paper Atlas

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

Spot the Difference: Detection of Topological Changes via Geometric Alignment

Steffen Czolbe, Aasa Feragen, Oswin Krause

NeurIPS 2021 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection · 0 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 semantic loss instead of MSE (ablation)Conditional VAE registration · Cell topological change detectionAUC: 0.72 -> 0.88Improved
Use semantic loss instead of MSE (ablation)Conditional VAE registration · Brain tumor detectionAUC: 0.61 -> 0.80Improved
Use the proposed semantic-loss method rather than Li and WyattUnsupervised topological change detection · Cell topological change detectionAUC: 0.75 -> 0.88Improved
Use the proposed semantic-loss method rather than Li and WyattUnsupervised brain tumor detection · Brain tumor detectionAUC: 0.70 -> 0.80Improved
Use the proposed semantic-loss method rather than supervised segmentationTopological change detection · Cell topological change detectionAUC: 0.90 -> 0.88Hurt
Use the proposed semantic-loss method rather than supervised segmentationBrain tumor detection · Brain tumor detectionAUC: 0.95 -> 0.80Hurt
Replace semantic loss with MSE (ablation)Conditional VAE registration · CellsDice: 0.90 -> 0.89Hurt
Replace semantic loss with MSE (ablation)Conditional VAE registration · BrainsDice: 0.56 -> 0.52Hurt
Replace semantic loss with MSE (ablation)Conditional VAE registration · CellsSeg. Accuracy (%): 95.4 -> 95.0Hurt
Replace semantic loss with MSE (ablation)Conditional VAE registration · BrainsSeg. Accuracy (%): 84.7 -> 84.1Hurt
Replace semantic loss with MSE (ablation)Conditional VAE registration · Brain tumor detectionAUC: 0.80 -> 0.61Hurt

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