Tight Integration of Feature-based Relocalization in Monocular Direct Visual Odometry

TL;DR

This paper presents a framework that combines feature-based relocalization with direct visual odometry to improve camera tracking accuracy in challenging environments.

Contribution

It introduces a novel integration of map-based relocalization into direct visual odometry, including pose priors, bundle adjustment, and an online fusion module.

Findings

Enhanced camera tracking accuracy in multi-weather conditions

Effective integration of handcrafted and learned features

Promising improvements demonstrated on real datasets

Abstract

In this paper we propose a framework for integrating map-based relocalization into online direct visual odometry. To achieve map-based relocalization for direct methods, we integrate image features into Direct Sparse Odometry (DSO) and rely on feature matching to associate online visual odometry (VO) with a previously built map. The integration of the relocalization poses is threefold. Firstly, they are incorporated as pose priors in the direct image alignment of the front-end tracking. Secondly, they are tightly integrated into the back-end bundle adjustment. Thirdly, an online fusion module is further proposed to combine relative VO poses and global relocalization poses in a pose graph to estimate keyframe-wise smooth and globally accurate poses. We evaluate our method on two multi-weather datasets showing the benefits of integrating different handcrafted and learned features and demonstrating promising improvements on camera tracking accuracy.