Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondences

TL;DR

This paper introduces MicKey, a novel keypoint matching pipeline that predicts metric 3D correspondences from image pairs, enabling accurate scale-aware relative pose estimation without depth data or scene reconstructions.

Contribution

MicKey is the first method to learn metric 3D correspondences directly from image pairs, eliminating the need for depth measurements or scene reconstructions during training.

Findings

Achieves state-of-the-art results on the Map-Free Relocalisation benchmark.

Requires less supervision than existing methods.

Does not rely on scene reconstructions or image overlap information.

Abstract

Given two images, we can estimate the relative camera pose between them by establishing image-to-image correspondences. Usually, correspondences are 2D-to-2D and the pose we estimate is defined only up to scale. Some applications, aiming at instant augmented reality anywhere, require scale-metric pose estimates, and hence, they rely on external depth estimators to recover the scale. We present MicKey, a keypoint matching pipeline that is able to predict metric correspondences in 3D camera space. By learning to match 3D coordinates across images, we are able to infer the metric relative pose without depth measurements. Depth measurements are also not required for training, nor are scene reconstructions or image overlap information. MicKey is supervised only by pairs of images and their relative poses. MicKey achieves state-of-the-art performance on the Map-Free Relocalisation benchmark while requiring less supervision than competing approaches.