LingGS: Decoupling Feed-Forward Geometry Prediction from Camera Tracking for High-Fidelity Gaussian SLAM

LingGS is an online monocular Gaussian splatting SLAM system that uses a frozen streaming 3D foundation model as a virtual RGB-D sensor for depth, then estimates camera motion with an explicit geometric solver based on feature reprojection. The authors report that this geometric tracking outperforms learned camera prediction, improves reconstruction quality on indoor and outdoor datasets, and runs at about 7 FPS on a single RTX 4090.