AI Paper Atlas

Research map / Image generation / GANs

Inclusive GAN: Improving Data and Minority Coverage in Generative Models

Ning Yu, Ke Li, Peng Zhou Jitendra Malik, Larry Davis, Mario Fritz

ECCV 2020 · Image generation › GANs · approach: GAN latent optimization and evaluation · 49 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 IMLE-GAN harmonized reconstruction to StyleGAN2StyleGAN2 · Stacked MNIST# modes: 940 -> 997Improved
add IMLE-GAN harmonized reconstruction to StyleGAN2StyleGAN2 · Stacked MNISTKL to uniform: 0.424 -> 0.200Improved
add IMLE-GAN harmonized reconstruction to StyleGAN2StyleGAN2 · CelebA valRecall30k: 0.741 -> 0.848Improved
add IMLE-GAN harmonized reconstruction to StyleGAN2StyleGAN2 · CelebA valFID30k: 9.49 -> 11.28Hurt
add IMLE-GAN harmonized reconstruction to StyleGAN2StyleGAN2 · CelebA valIvOM3k: 0.302 -> 0.262Improved
add interpolation-based augmentation (LPIPS) (ablation)Ours · CelebA valFID30k: 12.10 -> 11.28Improved
add interpolation-based augmentation (pixel) (ablation)Ours · CelebA valRecall30k: 0.763 -> 0.879Improved
use LPIPS distance instead of pixel distance (ablation)Ours · CelebA valFID30k: 34.46 -> 12.10Improved
use LPIPS distance instead of pixel distance (ablation)Ours · CelebA valRecall30k: 0.763 -> 0.843Hurt
train minority-specific model for Eyeglasses subgroup vs general modelOurs · CelebA Eyeglasses minority, valRecall1k: 0.708 -> 0.892Improved
train minority-specific model for Eyeglasses subgroup vs StyleGAN2StyleGAN2 · CelebA Eyeglasses minority, valIvOM1k: 0.352 -> 0.288Improved

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