A Real 3D Object Recognition Algorithm based on Virtual Training

C. de Trazegnies, J.P. Bandera, C. Urdiales, and F. Sandoval (Spain)


3D Object Recognition, Curvature Functions, Principal Component Analysis, Virtual training, Hidden Markov Models


This paper presents a new view based 3D object recognition system. The system relies on the sequentiality of a set of object views. These views are represented by short feature vectors extracted from the object shape curvature. Since the acquisition of all possible views of a real object is prob lematic, the system can be trained a priori with databases of virtual objects that are expected to be observed. Never theless, our system is capable of learning real objects in an unsupervised way. The proposed system has been success fully tested both for real and virtual objects.

Important Links:

Go Back