2014 | OriginalPaper | Buchkapitel
Optimizing an Artificial Dorsal Stream on Purpose for Visual Attention
verfasst von : Gustavo Olague, León Dozal, Eddie Clemente, Arturo Ocampo
Erschienen in: EVOLVE - A Bridge between Probability, Set Oriented Numerics, and Evolutionary Computation III
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Visual attention is a natural process performed by the brain, specifically by the dorsal stream, whose functionality is to perceive salient visual features. This chapter is devoted to the task of evolving an artificial dorsal stream (ADS) using the brain programming strategy. The idea is to state the problem of visual attention, normally studied as two parts: bottom-up and top-down, in terms of a unified approach following a teleological framework. Indeed, in this work visual attention is explained as a single mechanism that adapts itself according to a given task. In this way, brain programming is used to design ADSs. Experimental results show that this new approach can contrive ADSs useful in the solution of “top-down and bottom-up” visual attention problems. In particular, we present a solution to the size and missing pop-out problems that were unsolved previously in the literature.