AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

GRAM-HD: 3D-Consistent Image Generation at High Resolution with Generative Radiance Manifolds

Jianfeng Xiang, Jiaolong Yang, Yu Deng, Xin Tong

ICCV 2023 · 3D vision › NeRF and radiance fields · approach: 3D-aware generative radiance fields · 80 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
do super-resolution on radiance manifolds with 2D CNN instead of GRAM direct renderingGRAM · FFHQ 256FID: 15.0 -> 11.8Improved
use manifold super-resolution at 1024^2GRAM · FFHQ 1024 inferencerunning time: 6.69s -> 0.36sImproved
3D-consistent manifold super-resolution vs image-space upsampling methodsGRAM-HD vs StyleNeRF/StyleSDF/EG3D · FFHQ 1024 FIDFID: 9.45 -> 12.0Hurt
3D-consistent manifold super-resolution vs image-space upsampling methodsGRAM-HD vs StyleNeRF · FFHQ 1024 NeuS reconstructionPSNR: 30.0 -> 33.8Improved
add cross-resolution consistency loss L_cons (ablation)GRAM-HD · FFHQ 256FID: 16.9 -> 11.8Improved
add patch adversarial loss L_patch (ablation)GRAM-HD · FFHQ 256FID: 12.0 -> 11.8Improved
add dedicated background super-resolution network (ablation)GRAM-HD · FFHQ 256FID: 13.3 -> 12.0Improved
add LR generator intermediate features as input (ablation)GRAM-HD · FFHQ 256FID: 14.0 -> 13.3Improved
add sub-pixel convolutions (ablation)GRAM-HD · FFHQ 256FID: 14.4 -> 14.1Improved
use ESRGAN base network instead of StyleGAN2 decoder (ablation)super-resolution CNN · FFHQ 256FID: 18.2 -> 14.4Improved

Later papers that compared with this method

1 later accepted paper reports GRAM-HD as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 4 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