Linear and non-linear model for statistical localization of landmarks

B. Romaniuk, M. Desvignes, M. Revenu, M.J. Deshayes
Object recognition supported by user interaction for service robots  
This paper presents and compares 3 methods for the statistical localization of partially occulted landmarks. In many real applications, some information is visible in images and some parts are missing or occulted. These parts are estimated by 3 statistical approaches : a rigid registration, a linear method derived from PCA, which represents spatial relationships, and a non linear model based upon Kernel PCA. Applied to the cephalometric problem, the best method exhibits a mean error of 3.3 mm,
more » ... hich is about 3 times the intra-expert variability.
doi:10.1109/icpr.2002.1047478 dblp:conf/icpr/RomaniukDRD02 fatcat:pv2celqkkfevfl2gc6tm2kalya