AtlasGS: Atlanta-world Guided Surface Reconstruction with Implicit Structured Gaussians

TL;DR

This paper introduces AtlasGS, a novel surface reconstruction method that leverages Atlanta-world guidance and implicit structured Gaussian splatting to achieve smooth, detailed, and efficient 3D reconstructions of indoor and urban environments.

Contribution

The paper presents a new Atlanta-world guided implicit-structured Gaussian Splatting approach that improves global consistency and detail preservation in 3D scene reconstruction.

Findings

Outperforms state-of-the-art methods in indoor scene reconstruction.

Achieves high-frequency detail preservation and smoothness.

Demonstrates superior surface reconstruction quality in urban scenes.

Abstract

3D reconstruction of indoor and urban environments is a prominent research topic with various downstream applications. However, existing geometric priors for addressing low-texture regions in indoor and urban settings often lack global consistency. Moreover, Gaussian Splatting and implicit SDF fields often suffer from discontinuities or exhibit computational inefficiencies, resulting in a loss of detail. To address these issues, we propose an Atlanta-world guided implicit-structured Gaussian Splatting that achieves smooth indoor and urban scene reconstruction while preserving high-frequency details and rendering efficiency. By leveraging the Atlanta-world model, we ensure the accurate surface reconstruction for low-texture regions, while the proposed novel implicit-structured GS representations provide smoothness without sacrificing efficiency and high-frequency details. Specifically, we propose a semantic GS representation to predict the probability of all semantic regions and deploy a structure plane regularization with learnable plane indicators for global accurate surface reconstruction. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches in both indoor and urban scenes, delivering superior surface reconstruction quality.