Title :
Homography estimation using one ellipse correspondence and minimal additional information
Author :
Alvarez, Luis ; Caselles, Vicent
Author_Institution :
Dept. de Inf. y Sist., Univ. de Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Spain
Abstract :
In sport scenarios like football or basketball, we often deal with central views where only the central circle and some additional primitives like the central line and the central point or a touch line are visible. In this paper we first characterize, from a mathematical point of view, the set of homographies that project a given ellipse into the unit circle, next, using some extra minimal additional information like the knowledge of the position in the image of the central line and central point or a touch line we show a method to fully determine the plane homography. We present some experiments in sport scenarios to show the ability of the proposed method to properly recover the plane homography.
Keywords :
computational geometry; image processing; central circle; central line; central point; central views; ellipse correspondence; ellipse projection; homography estimation; minimal additional information; sport scenarios; touch line; unit circle; Calibration; Cameras; Equations; Estimation; Lenses; Mathematical analysis; Transforms; Homography estimation; camera calibration; ellipse correspondence;
Conference_Titel :
Image Processing (ICIP), 2014 IEEE International Conference on
Conference_Location :
Paris
DOI :
10.1109/ICIP.2014.7025981