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 treated as pose priors and tightly integrated into the direct
image alignment of the front-end tracking. Secondly, they are also tightly
integrated into the back-end bundle adjustment. 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.