High-Resolution Comprehensive 3-D Dynamic Database for Facial Articulation Analysis

Abstract

Face is an essential medium for complex non-verbal communication between humans. Its articulation plays an important role in everyday life, including: manifestation of emotional and awareness states, cognitive activity, personality or wellbeing. With the advances in 3-D imaging technology and ever increasing computing power, automatic analysis of facial articulation using 3-D sequences has become viable. This paper describes a newly developed high-resolution, comprehensive 3-D dynamic facial articulation database which is designed not only for comparative studies of facial expression analysis but also to aid research into clinical diagnosis of facial dysfunctions. The database currently contains 3,360 facial sequences captured from 80 volunteers of various age, gender and ethnicity and will be made available to the wider research community in early autumn 2011. It has been validated using computer-based experiments as well as psychophysical experiments that were used to formally evaluate the accuracy of the recorded expressions. This database is believed to be one of the most comprehensive repositories of facial 3D dynamic articulations to date.