Face Description and Object Analysis

The University of York has developed a unique method by which a wide variety of information can be estimated from digital images of objects. The technology is based on a statistical approach that can process information in real time (building on standard computer hardware and software platforms) to identify specific objects and make estimates of any number of visually apparent parameters.

The current embodiment of the system is designed to identify human faces in images then estimate the age, gender, ethnicity and facial expression of the subject. This has the potential for very significant commercial impact in the market research domain, and could be applied to a wide range of other image processing problems in other sectors, potentially including vehicle recognition.

Benefits

Automatic estimation of visually apparent parameters

Applications

  • Demographic analysis
  • Entertainment – games, art installations
  • Surveillance
  • Movie logging
  • Preprocessing for face recognition
  • Facial expression user interfaces
  • Portrait enhancement

Investment readiness status

The technology is developed for integration into certain applications, and requires additional technological development for others.

IP status

Many aspects of the core technology remain unpublished and therefore remain a trade secret
 

Contact details

Dr Mark Mortimer
Director of Research & Enterprise

Tel: 01904 435103