Attention in Stereo Vision: Implications for Computational Models of Attention

Attention in Stereo Vision: Implications for Computational Models of Attention

Neil D. B. Bruce (University of Manitoba, Canada) and John K. Tsotsos (York University, Canada)
DOI: 10.4018/978-1-4666-2539-6.ch004


The stereo correspondence problem is a topic that has been the subject of considerable research effort. What has not yet been considered is an analogue of stereo correspondence in the domain of attention. In this chapter, the authors bring this problem to light, revealing important implications for computational models of attention, and in particular, how these implications constrain the problem of computational modeling of attention. A model is described which addresses attention in the stereo domain, and it is revealed that a variety of behaviors observed in binocular rivalry experiments are consistent with the model’s behavior. Finally, the authors consider how constraints imposed by stereo vision may suggest analogous constraints in other non-stereo feature domains with significant consequence to computational models of attention.
Chapter Preview

2. The Need For Attention

Attention provides a mechanism for selection of particular aspects of a scene for subsequent processing while eliminating interference from competing visual events. A common misperception is that attention and fixation are one and the same phenomenon. Attention focuses processing on a selected region of the visual field that needn’t coincide with the centre of fixation. This is perhaps exemplified by the perceived ability to look out of the corner of one’s eye. There exist numerous formal arguments demonstrating the necessity of attention to solve the visual search problem (Tsotsos, 1988; Burt, 1988). In lieu of exhaustively describing each of these arguments, we instead summarize some of the more important elements and comment specifically on implications in the domain of stereo vision.

Complete Chapter List

Search this Book: