By Topic

Robust homography for real-time image un-distortion

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Jianhui Chen ; Dept. of Comput. Sci. & Eng., York Univ., Toronto, ON, Canada ; Benzeroual, K. ; Allison, R.S.

Stereoscopic 3D film production has increased the need for efficient and robust camera calibration and tracking. Many of these tasks involve making planar correspondence and thus accurate fast homography estimation is essential. However, homography estimation may fail with distorted images since the planar projected corners may be distorted far away from the “perfect” locations. On the other hand, precisely estimating lens distortion from a single image is still a challenge, especially in real-time applications. In this paper, we drop the assumption that the image distortion is negligible in homography estimation. We propose robust homography as a simple and efficient approach which combines homography mapping and image distortion estimation in a least square constraint. Our method can simultaneously estimate homography and image distortion from a single image in real-time. Compared with previous methods, it has two advantages: first, un-distortion can be achieved with little overhead due to the need for only a single calibration image and the real-time homography mapping of easy to track corners; second, due to the use of precise calibration targets the accuracy of our method is comparable to the multiple image calibration methods. In an experimental evaluation, we show that our method can accurately estimate image distortion parameters in both synthetic and real images. We also present its applications in close range un-distortion and robust corner detection.

Published in:

3D Imaging (IC3D), 2013 International Conference on

Date of Conference:

3-5 Dec. 2013