EpiDiffVO: Geometry-Aware Epipolar Diffusion for Robust Visual Odometry

TL;DR

EpiDiffVO introduces a geometry-aware, diffusion-based sparse matching framework with graph neural networks for robust visual odometry, reducing correspondence redundancy and improving pose estimation accuracy.

Contribution

The paper presents a novel sparse epipolar matching method combined with diffusion refinement and graph neural networks for end-to-end geometric pose estimation.

Findings

Reduces correspondence redundancy while maintaining accuracy.

Improves robustness of pose estimation across challenging baselines.

Effective on TartanAir and KITTI SLAM datasets.

Abstract

Estimating relative pose from image pairs fundamentally requires only a minimal subset of geometrically consistent correspondences. However, most learning-based approaches rely on dense matching or direct regression, leading to redundancy and reduced geometric interpretability. In this work, we propose a sparse epipolar matching framework that predicts a compact set of correspondences optimized for geometric consistency across varying temporal baselines. To address residual noise and misalignment, we introduce an epipolar diffusion process that models correspondence uncertainty and refines keypoints toward epipolar consistency. The refined correspondences, along with depth cues, are lifted into a graph representation forming a Steiner graph that encodes relational structure between points. A graph neural network learns a compact subset of informative correspondences, which are passed to a differentiable singular value decomposition solver for end-to-end geometric estimation. Relative pose is recovered from the resulting essential matrix and evaluated in a visual odometry setting on the TartanAir and KITTI SLAM datasets. Experimental results demonstrate that combining sparse matching, diffusion-based refinement, and graph-based subset selection reduces correspondence redundancy while maintaining robust pose estimation across challenging baselines.