NVSim: Novel View Synthesis Simulator for Large Scale Indoor Navigation

TL;DR

NVSim is a framework that creates large-scale indoor navigation simulators from simple image sequences, using innovative Gaussian Splatting techniques to ensure visual quality and navigability without extensive 3D scanning.

Contribution

It introduces Floor-Aware Gaussian Splatting and a mesh-free traversability algorithm for scalable indoor simulator construction from real-world data.

Findings

Successfully generates large-scale navigation graphs from real data

Addresses visual artifacts on sparse observations

Enables scalable indoor navigation simulation

Abstract

We present NVSim, a framework that automatically constructs large-scale, navigable indoor simulators from only common image sequences, overcoming the cost and scalability limitations of traditional 3D scanning. Our approach adapts 3D Gaussian Splatting to address visual artifacts on sparsely observed floors a common issue in robotic traversal data. We introduce Floor-Aware Gaussian Splatting to ensure a clean, navigable ground plane, and a novel mesh-free traversability checking algorithm that constructs a topological graph by directly analyzing rendered views. We demonstrate our system's ability to generate valid, large-scale navigation graphs from real-world data. A video demonstration is avilable at https://youtu.be/tTiIQt6nXC8