HI-SLAM: Monocular Real-Time Dense Mapping With Hybrid Implicit Fields

Wei Zhang, Tiecheng Sun, Sen Wang, Qing Cheng, Nobert Haala

Abstract:

In this letter, we present a neural field-based realtime monocular mapping framework for accurate and dense
Simultaneous Localization and Mapping (SLAM). Recent neural
mapping frameworks show promising results, but rely on RGBD or pose inputs, or cannot run in real-time. To address these
limitations, our approach integrates dense-SLAM with neural
implicit fields. Specifically, our dense SLAM approach runs
parallel tracking and global optimization, while a neural fieldbased map is constructed incrementally based on the latest
SLAM estimates. For the efficient construction of neural fields,
we employ multi-resolution grid encoding and signed distance
function (SDF) representation. This allows us to keep the map
always up-to-date and adapt instantly to global updates via
loop closing. For global consistency, we propose an efficient
Simp3q-based pose graph bundle adjustment (PGBA) approach
to run online loop closing and mitigate the pose and scale
drift. To enhance depth accuracy further, we incorporate learned
monocular depth priors. We propose a novel joint depth and scale
adjustment (JDSA) module to solve the scale ambiguity inherent
in depth priors. Extensive evaluations across synthetic and realworld datasets validate that our approach outperforms existing
methods in accuracy and map completeness while preserving
real-time performance.