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

Shoplyfter - — Natalia Queen

"Sophia, can you tell me where you were last night between 9 pm and 1 am?" Jameson asked, his eyes narrowing.

Natalia's eyes sparkled. "I've heard great things about your work, Detective. And I must admit, I was impressed by your sharp instincts. You're quite the sleuth."

Jameson's radar began to ping. He asked Natalia to bring Sophia in for questioning. Shoplyfter - Natalia Queen

Natalia's expression turned thoughtful. "Actually, I do have one employee who has been acting strange lately. Her name is Sophia, and she's been working for me for about six months. She's always been reliable, but I've noticed she's been distant and preoccupied lately."

The game was afoot. Jameson sensed that Sophia was hiding something, but he needed concrete evidence to tie her to the crime. "Sophia, can you tell me where you were

Natalia, a stunning woman in her late 20s with piercing green eyes and raven-black hair, arrived at the police station, her designer handbag slung over her shoulder. She was visibly shaken but determined to help the detective solve the case.

Jameson smiled. "It was my pleasure, Ms. Queen. But I have to ask – what made you think of me and not the other detectives?" And I must admit, I was impressed by your sharp instincts

"Please, Detective Jameson, you have to catch this person," Natalia begged, her voice trembling. "They've stolen my most prized possessions – a diamond-encrusted necklace and a pair of earrings worth over $50,000."

"I... I don't think so," Sophia stammered.

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