AI Paper Atlas

Research map / 3D vision / Gaussian splatting

PolGS: Polarimetric Gaussian Splatting for Fast Reflective Surface Reconstruction

Yufei Han, Bowen Tie, Heng Guo, Youwei Lyu, Si Li, Boxin Shi, Yunpeng Jia, Zhanyu Ma

ICCV 2025 · 3D vision › Gaussian splatting · approach: Regularization and primitive design for 3DGS · 5 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 polarimetric Stokes loss L_pol (ablation)PolGS · SMVP3D + RMVP3D meanmean angular error (deg): 14.89 -> 14.35Improved
add polarimetric Stokes loss L_pol (ablation)PolGS · SMVP3D + RMVP3D meanChamfer distance (mm): 7.80 -> 7.57Improved
add L_pol on texture-less David statue (ablation)PolGS · SMVP3D Davidmean angular error (deg): 15.44 -> 13.99Improved
add L_pol on Dragon (ablation)PolGS · SMVP3D Dragonmean angular error (deg): 25.74 -> 24.23Improved
add L_pol on Snail (ablation)PolGS · SMVP3D SnailChamfer distance (mm): 10.75 -> 10.85No clear effect
add L_pol for texture-rich surfaces with dense views (ablation)PolGS · texture-rich objects, >30 viewsreconstruction accuracyNo clear effect
add L_pol with fewer input views (ablation)PolGS · PISR texture-less real datareconstruction qualityImproved
PolGS instead of Gaussian SurfelsGaussian Surfels · SMVP3D + RMVP3D meanChamfer distance (mm): 10.30 -> 7.57Improved
PolGS instead of 3DGS-DR3DGS-DR · SMVP3D + RMVP3D meanmean angular error (deg): 19.76 -> 14.35Improved
PolGS (3DGS) instead of SDF-based NeRSP in training timeNeRSP · SMVP3D + RMVP3Dtraining time (hours): 10 h -> 0.1 hImproved
SDF-based NeRO versus PolGSPolGS · SMVP3D + RMVP3D meanmean angular error (deg): 14.35 vs 8.27 for NeROHurt

Later papers that compared with this method

1 later accepted paper reports PolGS as a baseline in their results tables; 0 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 0 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