Relational Epipolar Graphs for Robust Relative Camera Pose Estimation

TL;DR

This paper introduces a graph-based approach for robust relative camera pose estimation that leverages relational inference over epipolar correspondence graphs, improving robustness to noise and large baselines.

Contribution

It reformulates pose estimation as a relational inference problem over epipolar graphs, integrating classical and learning-based methods for enhanced accuracy.

Findings

Outperforms classical and learning-guided methods in noisy and large baseline scenarios.

Uses dense detector-free matching with LoFTR for improved keypoint correspondence.

Achieves more robust and accurate relative pose estimation on benchmark datasets.

Abstract

A key component of Visual Simultaneous Localization and Mapping (VSLAM) is estimating relative camera poses using matched keypoints. Accurate estimation is challenged by noisy correspondences. Classical methods rely on stochastic hypothesis sampling and iterative estimation, while learning-based methods often lack explicit geometric structure. In this work, we reformulate relative pose estimation as a relational inference problem over epipolar correspondence graphs, where matched keypoints are nodes and nearby ones are connected by edges. Graph operations such as pruning, message passing, and pooling estimate a quaternion rotation, translation vector, and the Essential Matrix (EM). Minimizing a loss comprising (i) $\mathcal{L}_2$ differences with ground truth (GT), (ii) Frobenius norm between estimated and GT EMs, (iii) singular value differences, (iv) heading angle differences, and (v) scale differences, yields the relative pose between image pairs. The dense detector-free method LoFTR is used for matching. Experiments on indoor and outdoor benchmarks show improved robustness to dense noise and large baseline variation compared to classical and learning-guided approaches, highlighting the effectiveness of global relational consensus.