M^3: Dense Matching Meets Multi-View Foundation Models for Monocular Gaussian Splatting SLAM

TL;DR

M^3 introduces a dense matching approach within a multi-view foundation model to significantly improve monocular SLAM accuracy and robustness in dynamic environments, achieving state-of-the-art results.

Contribution

It develops a novel M^3 framework that combines dense correspondence estimation with monocular SLAM, enhancing pose accuracy and scene reconstruction quality.

Findings

Reduces ATE RMSE by 64.3% compared to VGGT-SLAM 2.0

Outperforms ARTDECO by 2.11 dB in PSNR on ScanNet++

Achieves state-of-the-art accuracy in diverse benchmarks

Abstract

Streaming reconstruction from uncalibrated monocular video remains challenging, as it requires both high-precision pose estimation and computationally efficient online refinement in dynamic environments. While coupling 3D foundation models with SLAM frameworks is a promising paradigm, a critical bottleneck persists: most multi-view foundation models estimate poses in a feed-forward manner, yielding pixel-level correspondences that lack the requisite precision for rigorous geometric optimization. To address this, we present M^3, which augments the Multi-view foundation model with a dedicated Matching head to facilitate fine-grained dense correspondences and integrates it into a robust Monocular Gaussian Splatting SLAM. M^3 further enhances tracking stability by incorporating dynamic area suppression and cross-inference intrinsic alignment. Extensive experiments on diverse indoor and outdoor benchmarks demonstrate state-of-the-art accuracy in both pose estimation and scene reconstruction. Notably, M^3 reduces ATE RMSE by 64.3% compared to VGGT-SLAM 2.0 and outperforms ARTDECO by 2.11 dB in PSNR on the ScanNet++ dataset.