3D camera tracking from disparity images

Kiyoung Kim; Woontack Woo

doi:10.1117/12.632741

24 June 2005 3D camera tracking from disparity images

Kiyoung Kim, Woontack Woo

Proceedings Volume 5960, Visual Communications and Image Processing 2005; 596042 (2005) https://doi.org/10.1117/12.632741
Event: Visual Communications and Image Processing 2005, 2005, Beijing, China

Abstract

In this paper, we propose a robust camera tracking method that uses disparity images computed from known parameters of 3D camera and multiple epipolar constraints. We assume that baselines between lenses in 3D camera and intrinsic parameters are known. The proposed method reduces camera motion uncertainty encountered during camera tracking. Specifically, we first obtain corresponding feature points between initial lenses using normalized correlation method. In conjunction with matching features, we get disparity images. When the camera moves, the corresponding feature points, obtained from each lens of 3D camera, are robustly tracked via Kanade-Lukas-Tomasi (KLT) tracking algorithm. Secondly, relative pose parameters of each lens are calculated via Essential matrices. Essential matrices are computed from Fundamental matrix calculated using normalized 8-point algorithm with RANSAC scheme. Then, we determine scale factor of translation matrix by d-motion. This is required because the camera motion obtained from Essential matrix is up to scale. Finally, we optimize camera motion using multiple epipolar constraints between lenses and d-motion constraints computed from disparity images. The proposed method can be widely adopted in Augmented Reality (AR) applications, 3D reconstruction using 3D camera, and fine surveillance systems which not only need depth information, but also camera motion parameters in real-time.

Citation Download Citation

Kiyoung Kim and Woontack Woo "3D camera tracking from disparity images", Proc. SPIE 5960, Visual Communications and Image Processing 2005, 596042 (24 June 2005); https://doi.org/10.1117/12.632741

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available