Human Reader: A Paradigm for a Multimedia-Based Human Interface
NTT Human Interface Laboratories, Nippon Telegraph and Telephone, 1-2356, Take, Yokosuka, Kanagawa, 238-03 Japan
Received:February 8, 1995Accepted:February 17, 1995Published:June 20, 1995
Keywords:Human interface, Computer vision, Computer graphics, Multimedia
A paradigm for better human interface called Human Reader is introduced with references to computer vision (CV) and computer graphics (CG) research projects on human images at NTT Human Interface Laboratories. CV and CG are regarded as dual problems of visual information processing. Our research includes the recognition of face, detection of head direction and head motion, lip motion analysis, facial expressions analysis, detection of hand or finger positions and movements, 3D head model generation, synchronized acquisition of shape and color, and rendering realistic face images having various facial expressions with complex components such as hair.
Cite this article as:Y. Suenaga, “Human Reader: A Paradigm for a Multimedia-Based Human Interface,” J. Robot. Mechatron., Vol.7 No.3, pp. 204-208, 1995.Data files: