GRLoc: Geometric Representation Regression for Visual Localization

TL;DR

This paper introduces GRLoc, a geometric representation regression method for visual localization that explicitly predicts 3D scene geometry to improve pose estimation accuracy.

Contribution

The paper proposes a novel inverse approach to absolute pose regression that predicts geometric scene representations directly from images, enhancing robustness and accuracy.

Findings

Achieves state-of-the-art results on 7-Scenes and Cambridge Landmarks datasets.

Explicit geometric decoupling improves pose estimation performance.

Model generalizes better by incorporating geometric priors.

Abstract

Absolute Pose Regression (APR) has emerged as a compelling paradigm for visual localization. However, APR models typically operate as black boxes, directly regressing a 6-DoF pose from a query image, which can lead to memorizing training views rather than understanding 3D scene geometry. In this work, we propose a geometrically-grounded alternative. Inspired by novel view synthesis, which renders images from intermediate geometric representations, we reformulate APR as its inverse that regresses the underlying 3D representations directly from the image, and we name this paradigm Geometric Representation Regression (GRR). Our model explicitly predicts two disentangled geometric representations in the world coordinate system: (1) a raymap's directions to estimate camera rotation, and (2) a corresponding pointmap to estimate camera translation. The final camera pose is then recovered from these geometric components using a differentiable deterministic solver. This disentangled approach, which separates the learned visual-to-geometry mapping from the final pose calculation, introduces a strong geometric prior into the network. We find that the explicit decoupling of rotation and translation predictions measurably boosts performance. We demonstrate state-of-the-art performance on 7-Scenes and Cambridge Landmarks datasets, validating that modeling the inverse rendering process is a more robust path toward generalizable absolute pose estimation.