Identification of deceptive behavioral cues extracted from video

Abstract

This research project investigates a novel approach for deriving behavioral deception cues from videotaped interactions. Researchers utilized inputs extracted from video to construct a set of two-dimensional spatial features. The features for thirty-eight video interactions were then analyzed using discriminant analysis and logistic regression. Through this exploratory study, the team has identified a number of promising features that help discriminate deception from truth. The techniques explored hold promise for the creation of near real time systems for transportation security professionals.