LT-Gaussian: Long-Term Map Update Using 3D Gaussian Splatting for Autonomous Driving

2025-08-03

AI Summary

To address the low efficiency and high computational cost of long-term updating of 3D Gaussian Splatting (3D-GS) maps in autonomous driving, this paper proposes LT-Gaussian—a novel framework for incremental map maintenance. It pioneers the integration of 3D Gaussian lattices into incremental mapping, synergizing multi-modal reconstruction, LiDAR data stream-based structural change detection, and collaborative optimization with prior maps to enable local, parameter-level updates of Gaussians. Key contributions include: (i) a lightweight structural change detection module that precisely localizes scene modifications; and (ii) a geometry-semantic consistency-driven strategy for Gaussian point insertion, deletion, and reparameterization—eliminating the need for full-map reconstruction. We establish the first benchmark for 3D-GS map updating on nuScenes. Experiments show that LT-Gaussian reduces training time by 72% versus from-scratch reconstruction, improves PSNR by 2.1 dB, and preserves topological consistency and rendering fidelity.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Computer Vision: 3D Computer Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Maps play an important role in autonomous driving systems. The recently proposed 3D Gaussian Splatting (3D-GS) produces rendering-quality explicit scene reconstruction results, demonstrating the potential for map construction in autonomous driving scenarios. However, because of the time and computational costs involved in generating Gaussian scenes, how to update the map becomes a significant challenge. In this paper, we propose LT-Gaussian, a map update method for 3D-GS-based maps. LT-Gaussian consists of three main components: Multimodal Gaussian Splatting, Structural Change Detection Module, and Gaussian-Map Update Module. Firstly, the Gaussian map of the old scene is generated using our proposed Multimodal Gaussian Splatting. Subsequently, during the map update process, we compare the outdated Gaussian map with the current LiDAR data stream to identify structural changes. Finally, we perform targeted updates to the Gaussian-map to generate an up-to-date map. We establish a benchmark for map updating on the nuScenes dataset to quantitatively evaluate our method. The experimental results show that LT-Gaussian can effectively and efficiently update the Gaussian-map, handling common environmental changes in autonomous driving scenarios. Furthermore, by taking full advantage of information from both new and old scenes, LT-Gaussian is able to produce higher quality reconstruction results compared to map update strategies that reconstruct maps from scratch. Our open-source code is available at https://github.com/ChengLuqi/LT-gaussian.

Problem

• Efficiently updating 3D-GS maps for autonomous driving
• Detecting structural changes between old and new LiDAR data
• Improving reconstruction quality by leveraging old and new scene data

Innovation

• Multimodal Gaussian Splatting for scene reconstruction
• Structural Change Detection via LiDAR comparison
• Targeted Gaussian-Map updates for efficiency