Machine understanding of facial expression of pain

Maja Pantic, Leon J. M. Rothkrantz
2002 Behavioral and Brain Sciences  
An automated system for monitoring facial expressions could increase the reliability, sensitivity, and precision of the research on the relationship between facial signs and experiences of pain, and it could lead to new insights and diagnostic methods. This commentary examines whether the research on facial expression of pain, as reported by Williams, provides a sufficient basis for machine understanding of pain-associated facial expressions. Automatic analysis of facial expressions is rapidly
more » ... ecoming an area of intense interest in computer vision and artificial intelligence research communities. The major impulse to investigate the machine vision problems of detecting, tracking, and interpreting human facial expressions comes from the potential benefits that could accrue from these efforts. Automated systems that sense, process, and interpret human facial signals have important commercial potential; they seem to have a natural place in commercial products such as computer systems for video conferencing, video telephony, video surveillance, face and visual speech synthesis, and pervasive perceptual man-machine interfaces. Furthermore, monitoring and interpreting facial signals are important to lawyers, the police, and security agents, who are often interested in issues concerning deception and attitude. Finally, basic research that uses measures of facial behavior including behavioral science, medicine, neurology, and psychiatry, would reap substantial benefits from inexpensive, reliable, and rapid facial-expression measurement tools. Such tools could greatly advance the quality of research in these fields by providing an increased reliability, sensitivity, and precision of facial measurements, by shortening the time to conduct research that is now lengthy and laborious, and by enabling many more researchers, who are presently inhibited by its expense and complexity, to use facial measurements. It is this potential improvement of basic research, including the research on the relationship between facial expressions and experiences of pain, that forms our major motivation to discuss here whether the research reported by Williams provides a sufficient basis for machine understanding of pain-associated facial expressions.
doi:10.1017/s0140525x02360084 fatcat:xw7s7444anfarikzdhqhqnfliq