In this paper, we proposed a new deep learning based dense monocular SLAM method. Compared to existing methods, the proposed framework constructs a dense 3D model via a sparse to dense mapping using learned surface normals. With single view learned depth estimation as prior for monocular visual odometry, we obtain both accurate positioning and high quality depth reconstruction. The depth and normal are predicted by a single network trained in a tightly coupled manner.Experimental results show that our method significantly improves the performance of visual tracking and depth prediction in comparison to the state-of-the-art in deep monocular dense SLAM.
@article{arxiv.1903.09199,
title = {Sparse2Dense: From direct sparse odometry to dense 3D reconstruction},
author = {Jiexiong Tang and John Folkesson and Patric Jensfelt},
journal= {arXiv preprint arXiv:1903.09199},
year = {2019}
}
Comments
Accepted to ICRA 2019 (RA-L option), video demo available at https://www.youtube.com/watch?v=3pbSHX72JC8&t=22s