TransSplat: Unbalanced Semantic Transport for Language-Driven 3DGS Editing

TL;DR

TransSplat introduces a novel semantic transport framework for language-driven 3D Gaussian Splatting editing, explicitly modeling semantic correspondences to improve local editing accuracy and structural consistency.

Contribution

It formulates 3D editing as a multi-view unbalanced semantic transport problem, explicitly establishing semantic correspondences to enhance editing precision.

Findings

TransSplat outperforms existing methods in local editing accuracy.

It effectively suppresses erroneous edits in non-target regions.

The approach improves structural consistency in 3D scene editing.

Abstract

Language-driven 3D Gaussian Splatting (3DGS) editing provides a more convenient approach for modifying complex scenes in VR/AR. Standard pipelines typically adopt a two-stage strategy: first editing multiple 2D views, and then optimizing the 3D representation to match these edited observations. Existing methods mainly improve view consistency through multi-view feature fusion, attention filtering, or iterative recalibration. However, they fail to explicitly address a more fundamental issue: the semantic correspondence between edited 2D evidence and 3D Gaussians. To tackle this problem, we propose TransSplat, which formulates language-driven 3DGS editing as a multi-view unbalanced semantic transport problem. Specifically, our method establishes correspondences between visible Gaussians and view-specific editing prototypes, thereby explicitly characterizing the semantic relationship between edited 2D evidence and 3D Gaussians. It further recovers a cross-view shared canonical 3D edit field to guide unified 3D appearance updates. In addition, we use transport residuals to suppress erroneous edits in non-target regions, mitigating edit leakage and improving local control precision. Qualitative and quantitative results show that, compared with existing 3D editing methods centered on enhancing view consistency, TransSplat achieves superior performance in local editing accuracy and structural consistency.