3D surface reconstruction based on binocular vision

Author

Xuesheng Li ; Kaiyu Qin ; Ping Yao ; Jun Yu ; Wenjie Wu ; Lu Chen

Author_Institution

Sch. of Aeronaut. & Astronaut., Univ. of Electron. Sci. & Technol. of China, Chengdu, China

fYear

2014

fDate

3-6 Aug. 2014

Firstpage

1861

Lastpage

1865

Abstract

In this paper, a novel approach to represent human hand based on parallel binocular vision system is proposed. Firstly, object in the field of view of the two cameras is sliced into many cross-section regions by means of a number of two dimensional virtual slices. Secondly, an appropriate ellipse which approximate human hand profile curve is acquired. Then, 3D gesture surface reconstruction is completed, composed of sufficient ellipses. Experiments show the ability to reconstruct extremely clean and simple model from a stereo pair of images.

Keywords

cameras; computer vision; curve fitting; gesture recognition; image reconstruction; image representation; stereo image processing; 3D gesture surface reconstruction; cameras; cross-section regions; human hand profile curve; human hand representation; parallel binocular vision system; stereo image pair; sufficient ellipses; two dimensional virtual slices; Cameras; Conferences; Image reconstruction; Machine vision; Solid modeling; Surface reconstruction; Three-dimensional displays; 3D gesture surface reconstruction; human hand profile curve; parallel binocular vision;

fLanguage

English

Publisher

ieee

Conference_Titel

Mechatronics and Automation (ICMA), 2014 IEEE International Conference on

Conference_Location

Tianjin

Print_ISBN

978-1-4799-3978-7

Type

conf

DOI

10.1109/ICMA.2014.6885985

Filename

6885985