atlas

Research map / Image generation / GANs

InsetGAN for Full-Body Image Generation

CVPR 2022 · Image generation › GANs · approach: Large-scale GAN latent analysis · 52 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 joint optimization of a specialized face GAN and the full-body GANFull-body human face refinement · Generated humans vs unrefined imagesuser preference for refined results: 98%Helped
Replace joint latent optimization with CoModGAN face inpaintingFull-body human face refinement · Generated humans vs unrefined imagesuser preference for refined results: 7%Hurt
Apply face refinement with joint optimizationFull-body human generator · Full-body images, t=0.7FID: 26.67 -> 25.33Helped
Apply face refinement with joint optimizationFull-body human generator · Face crops, t=0.7FID: 27.14 -> 31.61Hurt
Apply face refinement with joint optimizationFull-body human generator · Full-body images, t=0.4FID: 71.90 -> 69.58Helped
Apply face refinement with joint optimizationFull-body human generator · Face crops, t=0.4FID: 66.61 -> 61.57Helped
Jointly optimize body and face latents instead of optimizing only the face latent (ablation)Face refinement · Face-body compositionseam continuityHelped
Use L1 rather than L2 losses (ablation)InsetGAN latent optimization · Image compositionvisual performanceHelped

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