Pose Estimation for Vehicle-mounted Cameras via Horizontal and Vertical Planes

TL;DR

This paper introduces two efficient solvers for vehicle-mounted camera egomotion estimation using affine correspondences, applicable to different plane orientations, and demonstrates their accuracy and speed on synthetic and real data.

Contribution

The paper presents novel linear solvers for egomotion estimation from a single affine correspondence, tailored for planes perpendicular or orthogonal to the ground, improving efficiency and accuracy.

Findings

Methods are highly efficient with small linear systems.

Achieve comparable or better accuracy than traditional algorithms.

Faster performance when integrated into robust estimators.

Abstract

We propose two novel solvers for estimating the egomotion of a calibrated camera mounted to a moving vehicle from a single affine correspondence via recovering special homographies. For the first class of solvers, the sought plane is expected to be perpendicular to one of the camera axes. For the second class, the plane is orthogonal to the ground with unknown normal, e.g., it is a building facade. Both methods are solved via a linear system with a small coefficient matrix, thus, being extremely efficient. Both the minimal and over-determined cases can be solved by the proposed methods. They are tested on synthetic data and on publicly available real-world datasets. The novel methods are more accurate or comparable to the traditional algorithms and are faster when included in state of the art robust estimators.