GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense Mapping

TL;DR

GeoRefine is an online depth refinement system that combines hybrid SLAM, self-supervised learning, and TSDF fusion to produce accurate dense maps from monocular sequences, demonstrating robustness and low error rates.

Contribution

It introduces a novel online depth refinement framework integrating learning-based priors, self-supervision, and global mapping for improved dense mapping accuracy.

Findings

Achieves as low as 5% absolute relative depth errors.

Demonstrates robustness across multiple public datasets.

Effectively integrates learning-based and self-supervised methods in online mapping.

Abstract

We present a robust and accurate depth refinement system, named GeoRefine, for geometrically-consistent dense mapping from monocular sequences. GeoRefine consists of three modules: a hybrid SLAM module using learning-based priors, an online depth refinement module leveraging self-supervision, and a global mapping module via TSDF fusion. The proposed system is online by design and achieves great robustness and accuracy via: (i) a robustified hybrid SLAM that incorporates learning-based optical flow and/or depth; (ii) self-supervised losses that leverage SLAM outputs and enforce long-term geometric consistency; (iii) careful system design that avoids degenerate cases in online depth refinement. We extensively evaluate GeoRefine on multiple public datasets and reach as low as $5\%$ absolute relative depth errors.