Direct Sparse Mapping
This work addresses a key limitation in visual SLAM for robotics and AR/VR by enabling persistent mapping, though it is incremental as it builds on existing PBA techniques.
The authors tackled the problem of scene reobservations in photometric bundle adjustment (PBA) by introducing DSM, a monocular visual SLAM system with a persistent map, achieving the most accurate results to date on the EuRoC dataset for a direct method.
Photometric bundle adjustment, PBA, accurately estimates geometry from video. However, current PBA systems have a temporary map that cannot manage scene reobservations. We present, DSM, a full monocular visual SLAM based on PBA. Its persistent map handles reobservations, yielding the most accurate results up to date on EuRoC for a direct method.