Triangle Splatting SLAM

2026-05-29

AI Summary

Existing dense RGB-D SLAM systems struggle to simultaneously achieve high geometric accuracy and real-time mesh generation, limiting their utility for downstream geometric tasks. This work proposes the first SLAM framework based on a differentiable triangle "soup," introducing Triangle Splatting into SLAM for the first time. By leveraging differentiable rendering, the method jointly optimizes camera poses and geometry reconstruction, while incorporating constrained Delaunay triangulation to produce connected, high-quality meshes in real time. Evaluated on the Replica and TUM-RGBD datasets, the approach achieves superior geometric reconstruction accuracy compared to existing methods, maintains competitive tracking performance, and enables novel capabilities such as real-time mesh editing and dynamic deformation.

Abstract

We present a dense RGB-D SLAM system using differentiable triangles as the 3D map representation. While 3D Gaussian Splatting has emerged as the leading method for novel-view synthesis, triangles remain the standard primitive for traditional rendering hardware, game engines, and downstream tasks requiring explicit geometry such as simulation, collision, and editing. Recent offline methods have demonstrated that an unstructured 'triangle soup' can be optimised into a photorealistic mesh via Delaunay triangulation across a set of posed images. Building upon this insight, we present the first dense SLAM system to employ Triangle Splatting to perform both tracking and mapping through online differentiable rendering of a triangle soup. The map can be converted into a connected mesh on-the-fly via restricted Delaunay triangulation, enabling new online capabilities such as mesh deformation and collision checking. On Replica and TUM-RGBD, our system outperforms baselines on 3D geometry, matches the camera-tracking accuracy, and enables online mesh-based scene editing.

Problem

• Triangle Splatting
• SLAM
• dense RGB-D
• mesh representation
• online mapping

Innovation

• Triangle Splatting
• Differentiable Rendering
• Dense SLAM
• Online Meshing
• Delaunay Triangulation