AI Paper Atlas

Research map / Image generation / GANs

Dark Side Augmentation: Generating Diverse Night Examples for Metric Learning

Albert Mohwald, Tomas Jenicek, Ondřej Chum

ICCV 2023 · Image generation › GANs · approach: GAN-based image manipulation · 9 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
train retrieval with GAN-generated night anchors instead of GeM baselineGeM VGG-16 · Tokyo 24/7mAP: 79.4 -> 90.2Improved
train retrieval with GAN-generated night anchors instead of GeM baselineGeM ResNet-101 · Tokyo 24/7mAP: 85.0 -> 92.2Improved
use synthetic GAN night images instead of real paired day-night dataGeM VGG-16 · Tokyo 24/7mAP: 87.0 -> 90.2Improved
add diverse anchor mining (ablation)HEDN GAN C · Tokyo 24/7mAP: 88.7 -> 88.8Improved
add diverse anchor mining (ablation)CLAHE C baseline · Tokyo 24/7mAP: 85.4 -> 85.9Improved
add CLAHE normalization (ablation)HEDN GAN · Tokyo 24/7mAP: 88.0 -> 88.7Improved
train HED edge detector jointly (HEDN) instead of frozen HED (ablation)EdgeMAC · Tokyo 24/7mAP: 79.5 -> 81.9Improved
add real night images (SfM-N/D) to GAN-generated training data (ablation)CycleGAN C D · Tokyo 24/7mAP: 90.2 -> 88.6No clear effect
use 50% instead of 25% night anchors (ablation)retrieval training · Tokyo 24/7mAPNo clear effect
replace GAN generator with simple LAB intensity inversion augmentation (ablation)GeM · retrieval average over Tokyo, ROxf, RParavg mAP: 70.0 -> 71.2 (far below trained generators)Mixed
translate night queries to day at inference with generatorretrieval at test time · Tokyo 24/7mAP: 39 and 52 mAP (substantially below baseline)Hurt
apply color-invariant CIConv to ResNet-101 GeMGeM ResNet-101 · Revisited OxfordmAP: 65.3 -> 62.0Hurt

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