Title :
Euclidean Invariant Snake for Joint Stereoscopic Segmentation
Author :
Charmiw, M.A. ; Derrode, S. ; Ghorbel, F.
Author_Institution :
GRIFT, Ecole Nationale des Sci. de l´´Informatique, Manouba
Abstract :
In this paper, we present the application of a recent method of snakes with geometric shape prior for the segmentation of a pair of calibrated stereo images. Contours in one image of the pair are found using those in the other image as reference. The initialization of the snakes algorithm on the second image is performed automatically by matching only four particular points. The method is then adapted for object motion tracking in mono and stereo video sequences. The outlines of the algorithm are detailed and some experimental results are given and commented
Keywords :
geometry; image matching; image motion analysis; image segmentation; image sequences; stereo image processing; Euclidean invariant snake; Fourier descriptors; calibrated stereo images; epilines; geometric shape; image contour; image matching; joint stereoscopic segmentation; object motion tracking; snakes algorithm; stereo video tracking; Active contours; Cameras; GSM; Image converters; Image segmentation; Laboratories; Shape; Stereo vision; Tracking; Video sequences; Fourier descriptors; SSnakes; epilines; invariant; shape prior; stereo video tracking; stereovisioin; video tracking;
Conference_Titel :
Information and Communication Technologies, 2006. ICTTA '06. 2nd
Conference_Location :
Damascus
Print_ISBN :
0-7803-9521-2
DOI :
10.1109/ICTTA.2006.1684523
