AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Tile-wise vs. Image-wise: Random-Tile Loss and Training Paradigm for Gaussian Splatting

Xiaoyu Zhang, Weihong Pan, Xiaojun Xiang, Hongjia Zhai, Liyang Zhou, Hanqing Jiang, Guofeng Zhang

ICCV 2025 · 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
replace image-wise loss with random-tile loss (RT-Loss) in 3DGS3D-GS · Mip-NeRF360PSNR: 28.69 -> 29.66Improved
replace image-wise loss with RT-Loss3D-GS · Deep BlendingPSNR: Scaffold-GS 30.21 vs Ours 30.15; 3D-GS 29.41 -> 30.15Mixed
apply random-tile training to Scaffold-GSScaffold-GS · Mip-NeRF360PSNR: 28.84 -> 29.86Improved
use RT-Loss on scenes with more training views (SMERF)3D-GS · SMERF Alameda at 4k itersPSNR: 17.62 -> 20.11Improved
apply RT-Loss to dynamic scene reconstruction4D-GS · Neural 3D VideoPSNR: 31.15 -> 31.74Improved
add RT-SSIM structure loss on random tiles (ablation)RT-Loss · KITCHENPSNR: 31.65 -> 32.35Improved
use RT-SSIM instead of image-wise SSIM during random tile training (ablation)RT-Loss · COURTHOUSEPSNR: 23.50 -> 24.15Improved
use Iteration-Count rather than Tile-Count densification (ablation)RT-Loss · KITCHENPSNR: 31.95 -> 32.35Improved
use image-wise batch training (B=4) instead of random tiles (ablation)3D-GS · KITCHENPSNR: 31.67 -> 31.22Hurt
increase number of views V for tile sampling (ablation)RT-Loss · COURTHOUSEPSNR: 23.76 -> 24.29 (V=2 to V=6)Improved
uniform instead of non-uniform tile sampling (ablation)RT-Loss · KITCHENPSNR: 32.30 -> 32.38No clear effect
use RT-Loss with random point initialization instead of SfM (ablation)3D-GS · KITCHENPSNR: 29.89 -> 31.22Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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