Attention plays an important role in human processing of sensory information as a mean of focusing resources toward the most important inputs at the moment. It has in particular been shown to be a key component of vision. In vision it has been argued that the attentional processes are crucial for dealing with the complexity of real world scenes. The problem has often been posed in terms of visual search tasks. It has been shown that both the use of prior task and context information - top-down influences - and favoring information that stands out clearly in the visual field - bottom-up influences - can make such search more efficient. In a generic scene analysis situation one presumably has a combination of these influences and a computational model for visual attention should therefore contain a mechanism for their integration. Such models are abundant for human vision, but relatively few attempts have been made to define any that apply to computer vision. In this article we describe a model that performs such a combination in a principled way. The system learns an optimal representation of the influences of task and context and thereby constructs a biased saliency map representing the top-down information. This map is combined with bottom-up saliency maps in a process evolving over time as a function over the input. The system is applied to search tasks in single images as well as in real scenes, in the latter case using an active vision system capable of shifting its gaze. The proposed model is shown to have desired qualities and to go beyond earlier proposed systems.
展开▼
