GS-Share: Enabling High-fidelity Map Sharing with Incremental Gaussian Splatting

2025-10-03

AI Summary

To address the challenge of constructing high-fidelity, incrementally updatable, and low-overhead 3D maps for autonomous driving and augmented reality, this paper proposes an anchor-based global mapping framework integrating 3D Gaussian splatting representation, generative virtual-view augmentation, and lightweight incremental encoding. The method leverages anchors to enforce cross-view consistency, employs synthesized virtual views to enhance out-of-distribution view extrapolation, and introduces an efficient incremental update mechanism to accommodate dynamic scene evolution. Experiments demonstrate significant improvements: +11% in PSNR, +22% in LPIPS, and +74% in Depth L1 accuracy, alongside a 36% reduction in map transmission bandwidth. The framework thus achieves a favorable trade-off among reconstruction fidelity, generalization capability, and communication efficiency.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures

Abstract

Constructing and sharing 3D maps is essential for many applications, including autonomous driving and augmented reality. Recently, 3D Gaussian splatting has emerged as a promising approach for accurate 3D reconstruction. However, a practical map-sharing system that features high-fidelity, continuous updates, and network efficiency remains elusive. To address these challenges, we introduce GS-Share, a photorealistic map-sharing system with a compact representation. The core of GS-Share includes anchor-based global map construction, virtual-image-based map enhancement, and incremental map update. We evaluate GS-Share against state-of-the-art methods, demonstrating that our system achieves higher fidelity, particularly for extrapolated views, with improvements of 11%, 22%, and 74% in PSNR, LPIPS, and Depth L1, respectively. Furthermore, GS-Share is significantly more compact, reducing map transmission overhead by 36%.

Problem

• Enabling high-fidelity 3D map sharing with compact representation
• Achieving continuous map updates with incremental Gaussian splatting
• Reducing transmission overhead while maintaining photorealistic reconstruction quality

Innovation

• Anchor-based global map construction technique
• Virtual-image-based map enhancement method
• Incremental Gaussian splatting update system