Reduced egomotion estimation drift using omnidirectional views

TL;DR

This paper introduces a method that combines omnidirectional and perspective camera views to reduce drift in egomotion estimation, significantly improving accuracy over longer sequences.

Contribution

It proposes a novel approach leveraging omnidirectional views to enhance egomotion estimation accuracy compared to traditional methods.

Findings

Improved accuracy in simulated experiments

Enhanced estimation stability in real-world tests

Reduced drift over extended image sequences

Abstract

Estimation of camera motion from a given image sequence becomes degraded as the length of the sequence increases. In this letter, this phenomenon is demonstrated and an approach to increase the estimation accuracy is proposed. The proposed method uses an omnidirectional camera in addition to the perspective one and takes advantage of its enlarged view by exploiting the correspondences between the omnidirectional and perspective images. Simulated and real image experiments show that the proposed approach improves the estimation accuracy.