Learning sequential visual attention control through dynamic state space discretization

Similar to humans and primates, artificial creatures like robots are limited in terms of allocation of their resources to huge sensory and perceptual information. Serial processing mechanisms used in the design of such creatures demands engineering attentional control mechanisms. In this paper, we present a new algorithm for learning top-down sequential visual attention control for agents acting in interactive environments. Our method is based on the key idea, that attention can be learned best in concert with visual representations through automatic construction and discretization of the visual state space. The tree representing the top-down attention is incrementally refined whenever aliasing occurs by selecting the most appropriate saccadic direction. The proposed approach is evaluated on action-based object recognition and urban navigation tasks, where obtained results support applicability and usefulness of developed saccade movement method for robotics.

Authors

• Ali Borji
• Majid Nili Ahmadabadi
• Babak Nadjar Araabi

Keywords

• State-space methods
• Automatic control
• Robot sensing systems
• Orbital robotics
• Layout
• Robotics and automation
• Object recognition
• Delay
• Decision making
• Humans
• Discretion
• State Space
• Visual Attention
• Visual Control
• Control Of Visual Attention
• Visual Representation
• Top-down Control
• Visual Space
• Saccadic Movements
• Top-down Attention
• Top-down Attentional Control
• Saccade Direction
• Eye Movements
• Codebook
• Visual Features
• Focus Of Attention
• Optimal Policy
• Goal State
• Perception Of Status
• SIFT Features
• Leaf Node
• Grid Position
• Physical Actions
• Visual Content
• Memory Items
• Robotic Platform
• Objective View