Active Vision

Active vision is a subset of the full attentional capabilities of a vision system.
However, most research makes assumptions that reduce or eliminate the need for attention:
Fixed camera systems negate the need for selection of visual field
Pre-segmentation eliminates the need to select a region of interest
‘Clean’ backgrounds ameliorate the segmentation problem
Assumptions about relevant features and the ranges of their values reduce their search ranges
Knowledge of task domain negates the need to search a stored set of all domains
Knowledge of which objects appear in scenes negates the need to search a stored set of all objects
Knowledge of which events are of interest negates the need to search a stored set of all events

The point is that the extent of the search space is seriously reduced before the visual processing takes place, and most often even before the algorithms for solution are designed! However, it is clear that in everyday vision, and certainly in order to understand vision, these assumptions cannot be made. 

Ruzena Bajcsy knew this when she wrote (R. Bajcsy, Active Perception vs Passive Perception, Proc. IEEE Workshop on Computer Vision: Representation and Control, Bellaire, MI, 1985): 

“Active sensing is the problem of intelligent control strategies applied to the data acquisition process which will depend on the current state of data interpretation including recognition.”

We have looked at several aspects of this, including:
 Active Binocular Camera Systems  
Active Visual Search
Active Object Recognition
Active Object Recognition
Active Recognition within a Bayesian Framework
theoretical results on the complexity of active vision strategies  ..Complexity Of Active Vision
theoretical results on behaviour-based robot control  ...Complexity of Behaviour-based Control