AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

WikiChurches: A Fine-Grained Dataset of Architectural Styles with Real-World Challenges

Björn Barz, Joachim Denzler

NeurIPS 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 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
Balance training classes by randomly undersampling each class (ablation)ResNet-50 pretrained on ImageNet-1k, 224×224 · WikiChurches-6 testbalanced accuracy: +1.5 percent pointsImproved
Increase input resolution from 224×224 to 448×448 (ablation)ResNet-50 pretrained on ImageNet-1k with balanced training · WikiChurches-6 testbalanced accuracy: +4.2 percent pointsImproved
Use ImageNet-21k pretraining instead of ImageNet-1k (ablation)TResNet-M, balanced training, 448×448 · WikiChurches-6 testbalanced accuracy: +3.6 percent pointsImproved
Use ImageNet-21k pretraining instead of ImageNet-1k (ablation)TResNet-M, balanced training, lower-resolution input · WikiChurches-6 testbalanced accuracy: +4.8 percent pointsImproved
Use ImageNet-21k pretraining instead of ImageNet-1k (ablation)TResNet-M with imbalanced training · WikiChurches-6 testbalanced accuracy: +1.4 percent pointsImproved

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