GeoGS-SLAM: Online Monocular Reconstruction Using Gaussian Splatting with Geometric Priors

2026-07-13

AI Summary

This work proposes an uncalibrated, online monocular dense reconstruction system based on 3D Gaussian splatting that jointly optimizes camera poses and scene representation without requiring depth input. Existing Gaussian splatting–based SLAM approaches either rely on external depth sensors or sacrifice original RGB information, degrading reconstruction quality. In contrast, this method uniquely integrates learned geometric priors with RGB observations to guide the online construction and optimization of Gaussians. Key technical contributions include a feedforward visual-geometry model, Gaussian primitive sampling, a photometric-geometric joint loss, a coarse-to-fine optimization strategy, and real-time loop closure. Experiments demonstrate that the proposed approach significantly outperforms current state-of-the-art methods on both indoor and outdoor benchmarks, achieving superior rendering fidelity and tracking accuracy while maintaining real-time performance.

Abstract

SLAM methods based on 3D Gaussian Splatting (3DGS) have demonstrated impressive tracking and mapping performance, but typically require additional geometric information from external depth sensors. Meanwhile, recent SLAM systems that leverage geometric priors from pre-trained feed-forward models enable real-time dense reconstruction, yet often discard original RGB information during optimization, thus degrading overall reconstruction quality. We present GeoGS-SLAM, an online monocular dense reconstruction system that combines the 3DGS-based map representation with learned geometric priors. Given uncalibrated RGB input, we first employ a feed-forward visual geometry model to predict camera and scene priors. The Gaussian scene map is then expanded by directly sampling Gaussian primitives from both RGB input and geometric priors. Camera poses and the scene map are jointly optimized through a coarse-to-fine strategy that minimizes both photometric and geometric losses. To ensure global consistency, we further incorporate online loop closure detection and pose graph optimization. Extensive experiments across indoor and outdoor benchmarks demonstrate that GeoGS-SLAM achieves superior rendering quality and tracking accuracy compared to state-of-the-art methods while maintaining online real-time performance. Project page: https://rlgao.github.io/geogs_slam.

Problem

• Monocular SLAM
• 3D Gaussian Splatting
• Geometric Priors
• Dense Reconstruction
• Real-time Performance

Innovation

• Gaussian Splatting
• Geometric Priors
• Monocular SLAM
• Online Reconstruction
• Loop Closure