AI Paper Atlas

Research map / 3D vision / Gaussian splatting

PaD-GS: Leveraging Distortion Map for Panoramic Gaussian Splatting

Yihang Xu, Qiulei Dong

ECCV 2026 · 3D vision › Gaussian splatting · approach: Regularization and primitive design for 3DGS · 0 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 distortion-aware Gaussian decoding (DAGD) (ablation)Panoramic 3DGS baseline · 360Roam BarPSNR: 22.29 -> 23.13Improved
add distortion-aware Gaussian decoding (DAGD) (ablation)Panoramic 3DGS baseline · Ricoh360 Cat-towerPSNR: 24.92 -> 25.36Improved
add distortion-aware opacity modulation (DAOM) with weight map M (ablation)Baseline + DAGD · 360Roam BarPSNR: 23.13 -> 23.40Improved
add distortion-aware opacity modulation (DAOM) with weight map M (ablation)Baseline + DAGD · Ricoh360 Cat-towerPSNR: 25.36 -> 25.75Improved
add distortion-aware weight map M to the loss (ablation)Baseline + DAGD + DAOM · 360Roam BarPSNR: 23.24 -> 23.40Improved
add distortion-aware weight map M to the loss (ablation)Baseline + DAGD + DAOM · Ricoh360 Cat-towerLPIPS: 0.194 -> 0.187Improved
add distortion-aware weight map M to the loss (ablation)Baseline + DAGD + DAOM · 360Roam BarLPIPS: 0.245 -> 0.230Improved
add DAGD (ablation)Panoramic 3DGS baseline · 360Roam BarLPIPS: 0.257 -> 0.240Improved
use PaD-GS instead of OmniGSOmniGS · 360Roam (11 indoor scenes, average)PSNR: 24.84 -> 26.02Improved
use PaD-GS instead of SPaGSSPaGS · Ricoh360 (12 outdoor scenes, average)PSNR: 26.20 -> 26.54Improved
use PaD-GS instead of OmniGS (speed cost)OmniGS · 360RoamFPS: 62.73 -> 29.87Hurt
use PaD-GS instead of OmniGS (training time)OmniGS · 360Roamtraining time (h): 0.68 -> 1.35Hurt

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