Catálogo de publicaciones - libros

Improving the game play in middle size league requires a very robust visual opponent and teammate detection system. Because games are highly dynamic, the detection system also has to be very fast. That both conditions are not necessarily contradictory is shown in this paper. The described multilevel approach documents, that the combination of a simple color based attention control and a subsequent neural object classification can be applied successfully in real world scenarios. The presented results indicate a very good overall performance regarding robustness, flexibility and computational needs.

A descriptor is presented for characterizing local image patches in a scale invariant manner. The descriptor is biologically-plausible in that the necessary computations are simple and local. Two different methods for robot visual homing based on this descriptor are also presented and tested. The first method utilizes the common technique of corresponding descriptors between images. The second method determines a home vector more directly by finding the stationary local image patch most similar between the two images. We find that the first method exceeds the performance of Franz et. al’s . No statistically significant difference was found between the second method and the warping method.