Identifying individuals in video by combining 'generative' and discriminative head models | IEEE Conference Publication | IEEE Xplore