AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Personalized Image Generation for Color Vision Deficiency Population

Shuyi Jiang, Daochang Liu, Dingquan Li, Chang Xu

ICCV 2023 · Image generation › Text-to-image diffusion · approach: Multi-subject personalization · 13 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 local contrast loss to StyleGAN (ablation)StyleGAN · Abstract Art, protan 40%Local contrast decay: 0.7639 -> 0.3784Improved
add color information loss to StyleGAN (ablation)StyleGAN · Abstract Art, protan 40%Hellinger distance: 0.0186 -> 0.0114Improved
combine local contrast and color information losses (ablation)CVD-GAN · Abstract Art, protan 40%Local contrast decay: 0.3784 -> 0.3355Improved
increase weight of disentanglement loss alpha (ablation)CVD-GAN · Abstract ArtFID vs disentanglementMixed
increase weight beta of CVD loss (ablation)CVD-GAN · image quality vs CVD metricsFID and CVD metricsMixed
generate CVD-oriented images directly instead of StyleGAN plus recolor post-processingCVD-GAN · Abstract Art, protan 100%Perceptual loss: StyleGAN 1.0350 -> 0.6606; Huang recolor 0.5333Mixed
add CVD losses to StyleGANStyleGAN · FlowersFID: 8.23 -> 18.93Hurt
add CVD losses to StyleGANStyleGAN · Abstract ArtFID: 14.35 -> 17.73Hurt
apply CVD-GAN at low severityCVD-GAN · Flowers, protan 20%Local contrast decay: 0.5937 -> 0.6829Hurt

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