AI Paper Atlas

Research map / 3D vision / Face and shape

Relightable Neural Human Assets From Multi-View Gradient Illuminations

Taotao Zhou, Kai He, Di Wu, Teng Xu, Qixuan Zhang, Kuixiang Shao, Wenzheng Chen, Lan Xu, Jingyi Yu

CVPR 2023 · 3D vision › Face and shape · approach: Generative textures for 3D humans · 20 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
Use depth-guided reprojection of captured normal and albedo maps instead of Instant-NGP volume rendering (ablation)Neural human asset G-buffer synthesis · Held-out captured viewsPSNR: 30.529 -> 30.868Improved
Use depth-guided reprojection of captured normal and albedo maps instead of Instant-NGP volume rendering (ablation)Neural human asset G-buffer synthesis · Held-out captured viewsSSIM: 0.968 -> 0.972Improved
Use depth-guided reprojection of captured normal and albedo maps instead of Instant-NGP volume rendering (ablation)Neural human asset G-buffer synthesis · Held-out captured viewsRMSE: 0.030 -> 0.029Improved
Use depth-guided reprojection of captured normal and albedo maps instead of Instant-NGP volume rendering (ablation)Neural human asset G-buffer synthesis · Held-out captured viewsMean angular error: 4.314° -> 4.296°Improved
Use depth-guided reprojection of captured normal and albedo maps instead of Instant-NGP volume rendering (ablation)Neural human asset G-buffer synthesis · Held-out captured viewsNormal error <5°: 79.891% -> 80.848%Improved
Use depth-guided reprojection of captured normal and albedo maps instead of Instant-NGP volume rendering (ablation)Neural human asset G-buffer synthesis · Held-out captured viewsNormal error <25°: 96.013% -> 96.293%Improved
Train relighting networks with UltraStage-rendered data rather than prior training dataSingle-image human relighting · Human relighting test imagesUser preference over RH: 96.6 %Improved
Train relighting networks with UltraStage-rendered data rather than prior training dataSingle-image human relighting · Human relighting test imagesUser preference over RW: 83.3 %Improved
Use photometric-stereo normal maps to guide geometry reconstruction instead of RGB supervision (ablation)Neural human geometry reconstruction · Captured human scenesFine-detail reconstructionImproved

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