ORB-SLAM based ICP and Optical Flow Combination Registration Algorithm in 3D Dense Reconstruction

Liya Duan, Renxia Wu, Xin Wang, Yinghe Liu, Yongjie Ma, Hui Fang

2018

Abstract

Dense 3D reconstruction is important for applications in many fields. The existing depth information based methods are typically constrained in their effective camera-object distance that should be from 0.4m to 4m. We present a combination of optical flow method and ICP method, and fuse this method and RGB-D camera ORB-SLAM to make full use of the color and depth data and refine the registration results that can achieve a more accurate dense 3D reconstruction. The experiment shows that this combination method is effective for the dense map acquisition.

Paper Citation

in Harvard Style

Duan L., Wu R., Wang X., Liu Y., Ma Y. and Fang H. (2018). ORB-SLAM based ICP and Optical Flow Combination Registration Algorithm in 3D Dense Reconstruction.In Proceedings of the 2nd International Conference on Intelligent Manufacturing and Materials - Volume 1: ICIMM, ISBN 978-989-758-345-2, pages 537-541. DOI: 10.5220/0007535205370541

in Bibtex Style

@conference{icimm18,
author={Liya Duan and Renxia Wu and Xin Wang and Yinghe Liu and Yongjie Ma and Hui Fang},
title={ORB-SLAM based ICP and Optical Flow Combination Registration Algorithm in 3D Dense Reconstruction},
booktitle={Proceedings of the 2nd International Conference on Intelligent Manufacturing and Materials - Volume 1: ICIMM,},
year={2018},
pages={537-541},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007535205370541},
isbn={978-989-758-345-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Intelligent Manufacturing and Materials - Volume 1: ICIMM,
TI - ORB-SLAM based ICP and Optical Flow Combination Registration Algorithm in 3D Dense Reconstruction
SN - 978-989-758-345-2
AU - Duan L.
AU - Wu R.
AU - Wang X.
AU - Liu Y.
AU - Ma Y.
AU - Fang H.
PY - 2018
SP - 537
EP - 541
DO - 10.5220/0007535205370541