Share Email Print

Proceedings Paper

Outdoor scene 3D reconstruction from multiple point cloud
Author(s): Hui Chen; Yan Feng; Jinzhong Sun; Jian Yang; Chenggang Cui
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper presents a 3D reconstruction method fusing structure from motion and laser scanning method. The precision of the 3D reconstruction is improved by combining point cloud obtained from different sensors. First, a scaled principal component analysis-iterative closest point (a scaled-PCA-ICP) algorithm is proposed to do the registration that can overcome the influence of large scale variance. Then, with the large scale factor, a recalculated registration center is proposed by region segmentation to realize the point cloud registration again. Finally, the two-view point clouds are robustly matched using the proposed optimization method to complete 3D color reconstruction of the outdoor large scenes. The proposed 3D reconstruction method is evaluated on the non-synchronous database of the real-world multi-view vision sequences obtained in an outdoor environment and laser scanner. The experiment results show improvements in both accuracy and efficiency.

Paper Details

Date Published: 6 May 2019
PDF: 9 pages
Proc. SPIE 11069, Tenth International Conference on Graphics and Image Processing (ICGIP 2018), 1106912 (6 May 2019); doi: 10.1117/12.2524421
Show Author Affiliations
Hui Chen, Shanghai Univ. of Electric Power (China)
Yan Feng, Shanghai Univ. of Electric Power (China)
Jinzhong Sun, Shanghai Univ. of Electric Power (China)
Jian Yang, Shanghai Univ. of Electric Power (China)
Chenggang Cui, Shanghai Univ. of Electric Power (China)

Published in SPIE Proceedings Vol. 11069:
Tenth International Conference on Graphics and Image Processing (ICGIP 2018)
Chunming Li; Hui Yu; Zhigeng Pan; Yifei Pu, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?