GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense 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.
Cite
Text
Ji et al. "GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense Mapping." Proceedings of the European Conference on Computer Vision (ECCV), 2022. doi:10.1007/978-3-031-19769-7_21Markdown
[Ji et al. "GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense Mapping." Proceedings of the European Conference on Computer Vision (ECCV), 2022.](https://mlanthology.org/eccv/2022/ji2022eccv-georefine/) doi:10.1007/978-3-031-19769-7_21BibTeX
@inproceedings{ji2022eccv-georefine,
title = {{GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense Mapping}},
author = {Ji, Pan and Yan, Qingan and Ma, Yuxin and Xu, Yi},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
year = {2022},
doi = {10.1007/978-3-031-19769-7_21},
url = {https://mlanthology.org/eccv/2022/ji2022eccv-georefine/}
}