AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

Injecting 3D Perception of Controllable NeRF-GAN into StyleGAN for Editable Portrait Image Synthesis

Jeong-gi Kwak, Yuanming Li, Dongsik Yoon, Donghyeon Kim, David Han, Hanseok Ko

ECCV 2022 · 3D vision › NeRF and radiance fields · approach: 3D-aware generative radiance fields · 22 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 pose loss to SURF-GAN training (ablation)SURF-GAN · FFHQ 128x128pose error (x10^-2): 2.69 -> 2.36Improved
add pose loss to SURF-GAN training (ablation)SURF-GAN · FFHQ 128x128FID: 44.56 -> 45.92Hurt
add progressive growing to SURF-GAN training (ablation)SURF-GAN · CelebA 64x64FIDImproved
SURF-GAN vs pi-GAN on FFHQ image qualitySURF-GAN · FFHQ 64x64 trainedFID: 47.12 -> 44.56Improved
SURF-GAN vs pi-GAN on CelebA pose accuracySURF-GAN · CelebA 64x64 trainedpose error (x10^-2): 5.81 -> 6.10Hurt
SURF-GAN vs pi-GAN on multi-view identity consistencySURF-GAN · CelebA 64x64 trainedidentity similarity: 0.65 -> 0.68Improved
inject SURF-GAN 3D control into StyleGAN2 vs CIPS-3D3D-controllable StyleGAN · FFHQ 256x256FID: 6.97 -> 4.72Improved
inject SURF-GAN 3D control into StyleGAN2 vs pi-GAN speed3D-controllable StyleGAN · FFHQ 256x256frames per second: 4 -> 72Improved
3D-controllable StyleGAN vs pi-GAN on pose accuracy3D-controllable StyleGAN · FFHQ 256x256pose error (x10^-2): 3.81 -> 4.24Hurt

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