Software to Read Facial Expressions

Software developer Noldus Information Technology has launched FaceReader, a product billed as 'the first automated software system capable of automatically detecting and analyzing human facial expressions'.

The system claims to recognise specific properties in facial images and interpret them according to 'the six fundamental human emotions' - happy, sad, angry, surprised, scared, and disgusted - and a 'neutral state'. It also automatically classifies faces according to gender, age, ethnicity, and facial hair, all 'without the need for markers, calibration or training.' The system promises applications for human-computer interaction, usability testing, behavioral science, medicine, communication and education.

Inventor Marten den Uyl, of development partner VicarVision bv, says the grail of a 'robust marker-free facial expression recognition tool' has been sought by many groups around the world. 'We are proud to be the first who have succeeded.'

The new functionality can be combined with The Observer XT, Noldus' tool for the collection, analysis, and presentation of observational data, tying facial expression information in with event logging data, captured computer screens, physiological signals such as heart rate, and more. For example, an eye-tracker could be used to find out which picture a person was looking at when showing the emotion 'disgusted'. It also promises new applications in 'affective computing', whereby other software programs can respond instantaneously to the emotional state of the user. [Watch out for PCs that cringe when you are ready to strike them, or sing to you 2001-style when you are about to hit Ctrl Alt Delete. Lip reading is next - Ed.]

Noldus has global headquarters in The Netherlands and offices across Europe and the United States. Web site: www.noldus.com.

All articles 2006-23 written and edited by Mel Crowther and/or Nick Thomas, 2024- by Nick Thomas, unless otherwise stated.