A database for facial behavioural analysis

Abstract

There is substantial interest in detection of human behaviour that may reveal people with deliberate malicious intent, who are engaging in deceit. Technology exists that is able to detect changes in facial patterns of movement and thermal signatures on the face. However, there is data deficiency in the research community for further study. Therefore this project aims to overcome the data deficiency in psychology study and algorithms development. A within-subjects design experiment was conducted, using immigration as a scenario for investigate participants in control and experimental conditions. A random sample of 32 volunteers were recruited, their age group is within 18 - 33. The study design required participants to answer questions on two topics, one as themselves and one as a predefined character. Data regarding visible and thermal images of facial movement and behaviour were collected. A rich FACS-coded database with high quality thermal images was established. Finally, recommendations for development and subsequent implementation of the facial analysis technique were made.