Share Email Print

Optical Engineering

3-D reconstruction using image sequences based on projective depth and simplified iterative closest point
Author(s): Keju Peng; Xin Chen; Dongxiang Zhou; Yongping Zhai; Yunhui Liu
Format Member Price Non-Member Price
PDF $20.00 $25.00

Paper Abstract

The authors propose a new algorithm for 3-D construction from a sequence of images based on a new corollary that the projective depth can be comprehended as a scalar factor between two 3-D points when reconstructed from three images. This algorithm constructs the partial models of an object using two consecutive images in the image sequence and integrates the obtained partial models to a complete one on the basis of this new corollary. In order to avoid accumulation of errors in the integration process, this algorithm modifies the reconstruction results based on a simplified Iterative Closest Point (ICP) algorithm. We have carried out two groups of experiments based on images captured from a library environment. In one group of experiments, sparse points are used to reconstruct regular objects; in the other group of experiments, dense points are employed. We compared experimental results of the proposed algorithm with the optimization method using the fundamental matrix, which demonstrated that the proposed algorithm yielded better efficiency and accuracy of the 3-D reconstruction. Experiments also showed that the reconstruction errors of the proposed method were within 5%.

Paper Details

Date Published: 6 March 2012
PDF: 11 pages
Opt. Eng. 51(2) 021110 doi: 10.1117/1.OE.51.2.021110
Published in: Optical Engineering Volume 51, Issue 2
Show Author Affiliations
Keju Peng, National Univ. of Defense Technology (China)
Xin Chen, National Univ. of Defense Technology (China)
Dongxiang Zhou, National Univ. of Defense Technology (China)
Yongping Zhai, National Univ. of Defense Technology (China)
Yunhui Liu, The Chinese Univ. of Hong Kong (Hong Kong, China)

© SPIE. Terms of Use
Back to Top