Hippocampal Anatomy Supports the Use of Context in Object Recognition: A Computational Model

Technology #ua14-134

Questions about this technology? Ask a Technology Manager

Download Printable PDF

Categories
Researchers
Ragan Bhaltacharyya
Mike Howard
Patrick Greene
Graduate Associate, Mathematics
Jean-Marc Fellous
Associate Professor, Psychology
Managed By
John Geikler
Asst. Director, Physical Science Licensing (520) 626-4605

Title: Novel Object and Context Recognition Computational Methods based on Hippocampal Anatomy

 

Invention: Inspired by hippocampal anatomy, this invention stores both objects and contexts separately in a computational framework. In addition, it allows both sets of data to influence each other which in turn allows guesses to be made about data whether given either context or object. 

Background: Current options for computation for use in visual systems are based around object oriented computing. Object oriented computing is based on a hierarchical system where relationships between objects are identified and also stored in discrete levels (Kowalski). This only permits associations to be made between objects.

 

Applications:

  • Medical imaging devices

 

Advantages:

  • Faster and more meaningful associations between objects and contexts
  • Increased potential for basic learning capabilities
  • Increased accuracy and precision for object or context associations