Feature-based Recursive Observer Design for Homography Estimation

TL;DR

This paper introduces a recursive observer algorithm for real-time homography estimation in robotic vision, leveraging group theory and sensor data to enhance robustness under challenging conditions.

Contribution

It develops a novel recursive observer exploiting the Special Linear group structure combined with gyroscope and feature data for improved homography estimation.

Findings

Robust performance during fast camera motion

Effective under severe occlusion

Handles specular reflections well

Abstract

This paper presents a new algorithm for online estimation of a sequence of homographies applicable to image sequences obtained from robotic vehicles equipped with vision sensors. The approach taken exploits the underlying Special Linear group structure of the set of homographies along with gyroscope measurements and direct point-feature correspondences between images to develop temporal filter for the homography estimate. Theoretical analysis and experimental results are provided to demonstrate the robustness of the proposed algorithm. The experimental results show excellent performance even in the case of very fast camera motion (relative to frame rate), severe occlusion, and in the presence of specular reflections.