Title :
3D Pose Estimation of Cactus Leaves using an Active Shape Model
Author :
Moeslund, Thomas B. ; Aagaard, Michael ; Lerche, Dennis
Author_Institution :
Lab. of Comput. Vision & Media Technol., Aalborg Univ.
Abstract :
Future machine vision applications have to cope with 3D non-rigid objects. One such application is 3D pose estimation of cactus leaves, which is the topic in this paper. We apply an active shape model (ASM) to estimate the 2D pose of a leaf in one image. We post-process the ASM in order to find well-defined characteristics which are located in a second image using correlation. This process provides 3D points which are used to find the 3D pose of the leaf. Tests show that 84.6% of the 3D poses are found correctly
Keywords :
computer vision; correlation methods; 3D pose estimation; active shape model; cactus leaves; machine vision; Active shape model; Application software; Belts; Computer vision; Deformable models; Electronic mail; Laboratories; Machine vision; Prototypes; Testing;
Conference_Titel :
Application of Computer Vision, 2005. WACV/MOTIONS '05 Volume 1. Seventh IEEE Workshops on
Conference_Location :
Breckenridge, CO
Print_ISBN :
0-7695-2271-8
DOI :
10.1109/ACVMOT.2005.1
