3D object porototype-based recognition from 2D images using features net
Raquel César, Agostinho C. Rosa · International Conference on Signal Processing · 2005
Object recognition is at the top of a visual task hierarchy. In its general form, this is a very difficult computational problem, which will probably play an important role in the eventual building of intelligent machines. A large number of psychological and neurophysiologic studies support the idea that humans represent three-dimensional objects internally as a small set of bidimensional images. In this work we present a scheme for recognition of 3D objects from 2D images. The proposed approach begins by identifying the class of the observed object and only then proceeds to determine its individual identity. In this way, we are able to reduce the computational costs of an exhaustive comparison with all known objects. The developed system has no previous knowledge about existing objects and builds the object basis as it operates on given images.