MCGS-SLAM

A Multi-Camera SLAM Framework Using Gaussian Splatting for High-Fidelity Mapping

Anonymous Author
Code and data [Available Upon Acceptance]

SLAM System Pipeline

Our method performs real-time SLAM by fusing synchronized inputs from a multi-camera rig into a unified 3D Gaussian map. It first selects keyframes and estimates depth and normal maps for each camera, then jointly optimizes poses and depths via multi-camera bundle adjustment and scale-consistent depth alignment. Refined keyframes are fused into a dense Gaussian map using differentiable rasterization, interleaved with densification and pruning. An optional offline stage further refines camera trajectories and map quality. The system supports RGB inputs, enabling accurate tracking and photorealistic reconstruction.

Right Image

Analysis of Single-Camera and Multi-Camera System

This experiment on the Waymo Open Dataset (Real World) demonstrates the effectiveness of our Multi-Camera Gaussian Splatting SLAM system. We evaluate the 3D mapping performance using three individual cameras, Front, Front-Left, and Front-Right, and compare these single-camera reconstructions against the Multi-Camera SLAM results.

The comparison highlights that the Multi-Camera SLAM leverages complementary viewpoints, providing more complete and geometrically consistent 3D reconstructions. In contrast, single-camera setups are prone to occlusions and limited fields of view, resulting in incomplete or distorted geometry. Our approach effectively fuses information from all three perspectives, achieving superior scene coverage and depth accuracy.

Right Image

Ashurbanipal Here

He picked up a reed stylus, his mind drifting from the administrative tallies of grain and captured gold. He wanted to record something that would outlast the stone walls of his palace. He began to write the story of his own secret education.

Suddenly, a heavy curtain parted. A breathless messenger knelt on the floor, breaking the King's reverie. The messenger delivered news of another rebellion stirring in the south, in the ancient city of Babylon. ashurbanipal

Ashurbanipal did not rage or call for his generals. Instead, he looked down at the tablet in his hand. He realized that his true power did not lie in the iron tips of his army's spears, but in the vast, accumulated knowledge surrounding him in the dark. He knew the history of Babylon's past rebellions, the strategies of their former kings, and the psychological fractures of their people. He picked up a reed stylus, his mind

He set down his stylus and smiled at the messenger. He would march to war at dawn, but he would win the war tonight, right here among his books. He was not just a king of men, but the first true king of information. Suddenly, a heavy curtain parted

Analysis of Single-Camera and Multi-Camera SLAM (Tracking)

In this section, we benchmark tracking accuracy across eight driving sequences from the Waymo dataset (Real World). MCGS-SLAM achieves the lowest average ATE, significantly outperforming single-camera methods.
Right Image

We further evaluate tracking on four sequences from the Oxford Spires dataset (Real World). MCGS-SLAM consistently yields the best performance, demonstrating robust trajectory estimation in large-scale outdoor environments.
Right Image

Right Image