A New Non-Linear Algorithm for 3D Reconstruction from Straight-line Optical Flow

Author

Chen Zhen ; Li Ming ; Jiang Shao-feng ; Wei Shui-gen ; Wang Yu

Author_Institution

Lab. of Comput. Vision, Nanchang Hangkong Univ., Nanchang

fYear

2009

fDate

1-2 Feb. 2009

Firstpage

311

Lastpage

316

Abstract

This paper presents a new mathematical definition of straight-line optical flow. By using 21 straight-line optical flows in 3 consecutive image frames, a set of equations about 3D motion parameters are established to obtain 12 motion parameters of 3D object and coordinates of the 3D straight line. Usually, it is difficult to find the 21 straight lines on correspondence. In this paper, a minimum of 6 straight-line optical flows in 3 consecutive image frames is required to obtain 12 motion parameters of 3D object and coordinates of the 3D straight line, by solving non-linear equations. With the parameters and coordinates a 3D scene can be reconstructed.

Keywords

image motion analysis; image reconstruction; image sequences; 3D motion; 3D object; 3D reconstruction; 3D straight line; image frames; straight-line optical flow; Asia; Computer vision; Image motion analysis; Image reconstruction; Layout; Motion estimation; Nonlinear equations; Nonlinear optics; Robotics and automation; Spatiotemporal phenomena; Computer vision; Image sequence; Non-linear algorithm; Reconstructing 3D motion and structure; Straight-line optical flow;

fLanguage

English

Publisher

ieee

Conference_Titel

Informatics in Control, Automation and Robotics, 2009. CAR '09. International Asia Conference on

Conference_Location

Bangkok

Print_ISBN

978-1-4244-3331-5

Type

conf

DOI

10.1109/CAR.2009.98

Filename

4777248