GSVisLoc: Generalizable Visual Localization for Gaussian Splatting Scene Representations

TL;DR

GSVisLoc is a novel visual localization method that leverages 3D Gaussian Splatting scene representations to accurately estimate camera poses without retraining or additional data, demonstrating strong performance across diverse scenes.

Contribution

The paper introduces GSVisLoc, a generalizable localization approach that uses 3D Gaussian Splatting for scene representation, enabling accurate pose estimation without retraining.

Findings

Achieves competitive localization accuracy on standard benchmarks.

Outperforms existing 3DGS-based localization baselines.

Generalizes effectively to unseen scenes without retraining.

Abstract

We introduce GSVisLoc, a visual localization method designed for 3D Gaussian Splatting (3DGS) scene representations. Given a 3DGS model of a scene and a query image, our goal is to estimate the camera's position and orientation. We accomplish this by robustly matching scene features to image features. Scene features are produced by downsampling and encoding the 3D Gaussians while image features are obtained by encoding image patches. Our algorithm proceeds in three steps, starting with coarse matching, then fine matching, and finally by applying pose refinement for an accurate final estimate. Importantly, our method leverages the explicit 3DGS scene representation for visual localization without requiring modifications, retraining, or additional reference images. We evaluate GSVisLoc on both indoor and outdoor scenes, demonstrating competitive localization performance on standard benchmarks while outperforming existing 3DGS-based baselines. Moreover, our approach generalizes effectively to novel scenes without additional training.