AI Paper Atlas

Research map / 3D vision / Gaussian splatting

RobustSplat: Decoupling Densification and Dynamics for Transient-Free 3DGS

Chuanyu Fu, Yuqi Zhang, Kunbin Yao, Guanying Chen, Yuan Xiong, Chuan Huang, Shuguang Cui, Xiaochun Cao

ICCV 2025 · 3D vision › Gaussian splatting · approach: Regularization and primitive design for 3DGS · 16 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 transient mask MLP to 3DGS (ablation)3DGS · NeRF On-the-go SpotPSNR: 18.54 -> 24.75Improved
add delayed Gaussian growth to 3DGS+Mask+MB (ablation)RobustSplat · NeRF On-the-go SpotPSNR: 25.25 -> 26.21Improved
add scale-cascaded mask bootstrapping to 3DGS+Mask+DG (ablation)RobustSplat · NeRF On-the-go SpotPSNR: 25.61 -> 26.21Improved
add delayed Gaussian growth to plain 3DGS (no mask) (ablation)3DGS · NeRF On-the-go SpotPSNR: 18.54 -> 23.42Improved
add mask regularization (ablation)RobustSplat · NeRF On-the-go SpotPSNR: 25.63 -> 26.21Improved
add image residual loss to mask supervision (ablation)RobustSplat · NeRF On-the-go meanPSNR: 23.77 -> 24.54Improved
add feature cosine similarity loss to mask supervision (ablation)RobustSplat · NeRF On-the-go meanPSNR: 23.46 -> 24.54Improved
use DINOv2 features instead of SAM2 for mask MLP input (ablation)RobustSplat · NeRF On-the-go meanPSNR: 24.01 -> 24.54Improved
use DINOv2 features instead of Stable Diffusion for mask MLP input (ablation)RobustSplat · NeRF On-the-go meanPSNR: 24.04 -> 24.54Improved
RobustSplat versus T-3DGS on NeRF On-the-goT-3DGS · NeRF On-the-go meanPSNR: 22.87 -> 23.22Improved
RobustSplat versus T-3DGS on RobustNeRFT-3DGS · RobustNeRF meanPSNR: 28.25 -> 29.36Improved
RobustSplat versus T-3DGS on On-the-go IIT-3DGS · NeRF On-the-go II meanPSNR: 22.63 -> 23.02Improved

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