Papers by Author: Yong Yan Yu

Abstract: Be dead against the fact that most of algorithms are incapable to solve Multi-Models Estimation effectually,a new method is provided that estimate multi-models with Model-Based Clustering in Conceptual Space.Start from that each data point is represented with a Preference Set of hypotheses models preferred by that poin,and then,the Jaccard distance between two Preference Sets is described as a attribute of a data point, finally, to perform a clustering operation using the improved Cobweb algorithm based on the attribute of the data points.Neither this method requires prior specification of the number of models, nor it necessitate parameters transform,so that it can overcome missing detection and false detection of crossing models.

Abstract: Aimed at the fact that most of the objects was symmetrical, we present a method reconstructing 3D model from single 2D image. At first, the angle, inclination angle and pivot angle should be established in the ternary perspective transformed matrix T, then the calibration which indicated the outline of the object can pick up the characteristic line. It resolved the host extinguishes information in accordance with the similar features of the parallel group of lines projection angle, then ensured the viewpoint situation and the object plane of symmetry, and made use of the object imagination plane of symmetry in accordance with symmetry feature through the interactive alignment specifying three pairs of the known symmetry image coordinates and the space coordinates of the corresponding points confirm the perspective transformative matrix T, then reverse prove the other space coordinates of the surface of the objects.