In order to obtain knowledge regarding damage mechanisms of building located above surface of the 2016 Kumamoto earthquake fault, the residual deformations of the building were investigated with 3D point cloud data obtained by a laser scanner. The 3D point cloud data contain noise due to measurement error and diffuse reflection. Residual deformations were grasped accurately with method reducing the influence of measurement error by using RANSAC algorism which is one of robust estimations. The validity of the obtained results is also confirmed by comparing other research and evaluating the error.
