AI Paper Atlas

Research map / 3D vision / 3D generation

Stanford-ORB: A Real-World 3D Object Inverse Rendering Benchmark

Zhengfei Kuang, Yunzhi Zhang, Hong-Xing Yu, Samir Agarwala, Shangzhe Wu, Jiajun Wu

NeurIPS 2023 · 3D vision › 3D generation · approach: Monocular 3D object reconstruction · 10 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 IDR (SDF surface) instead of NeRFobject inverse rendering · Stanford-ORB novel view synthesisPSNR-H: 26.31 -> 30.11Improved
use IDR instead of NeRFobject inverse rendering · Stanford-ORB geometryChamfer distance x1e-3: 62.05 -> 0.30Improved
extend IDR/NeRF with material and lighting decompositionIDR/NeRF · Stanford-ORB novel view synthesisPSNR-H: 30.11 -> 25.91 (InvRender)Hurt
add Monte Carlo differentiable renderer (NVDiffRecMC vs NVDiffRec)NVDiffRec · Stanford-ORB novel view synthesisPSNR-H: 21.94 -> 28.03Improved
add Monte Carlo differentiable renderer (NVDiffRecMC vs NVDiffRec)NVDiffRec · Stanford-ORB novel scene relightingPSNR-H: 22.91 -> 24.43Improved
train with ground-truth 3D scans and pseudo materialsNVDiffRecMC · Stanford-ORB novel scene relightingPSNR-H: 24.43 -> 25.08Improved
add Monte Carlo renderer when ground-truth scans are usedNVDiffRec with GT scans · Stanford-ORB novel scene relightingPSNR-H: 24.93 -> 25.08No clear effect
change object appearance (dark Salt can vs Baking can of similar shape)NVDiffRecMC · Stanford-ORB novel scene relightingPSNR-H: 24.4 -> 7.5Hurt

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