Communications of the ACM
Natural 'barcodes' Help US Recognize Faces
University College London
Humans recognize faces by organizing facial features such as eyebrows, eyes, and lips into horizontal lines of information, say Dr. Steven Dakin of the University College London (UCL) Institute of Ophthalmology and professor Roger Watt from the University of Stirling. Their research has the potential to improve face recognition software and CCTV cameras.
The researchers manipulated the images of celebrities and found that their facial features could be rendered in horizontal stripes of information that are similar to the barcodes used on commercial products. Dakin and Watt also studied other natural images, such as flowers and landscapes, but found that faces are unique in conveying useful information this way. They say the barcode pattern is recognized efficiently by the visual parts of the brain, is easy to locate in complex scenes, and may be resistant to changes in the overall appearance of the face.
"To improve face recognition software, we need to look towards biology and see how we have solved the problem," Dakin says. "If we are looking for barcode-like images to tell us that 'this is a face,' then software could be developed to mimic this skill."
From UCL News
View Full Article
Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA 
No entries found