Science software could soon be able to emulate the way the brain visualises objects, thanks to the creation of a new computational model.
The model was created by scientists at the Massachusetts Institute of Technology McGovern Institute for Brain Research based on the way human perception is believed to operate.
Initially, the brain assesses an entire scene, identifying areas which it believes are potentially of the greatest interest.
Should a particular object's location be needed, these high-interest areas are examined first to see if it is there.
The
science software also takes into account the fact that identifying an object and determining where it is are carried out in two different areas of the brain.
Lead author on the project Sharat Chikkerur says: "The model that we have tries to explain how this information is integrated."
Mr Chikkerur is a fifth-year PhD candidate whose studies encompass neuroscience, computer vision, biometrics and machine learning.
Product directory
.jpg)
-(1).jpg)
