Fig 2.tif (1.9 MB)
Registration examples on the Chinese character point set.
dataset
posted on 2016-02-11, 12:47 authored by Lei Peng, Guangyao Li, Mang Xiao, Li XieFrom top to bottom are the four largest degradations: deformation (0.08), noise (0.05), occlusion (0.5), and outlier (2.0). The model point set is shown in blue pluses, and the scene point set is shown in red circles.