Image(s): 640*480

Jpeg Image (20 Ko)

Jpeg Image (37 Ko)

Author(s)

• Keith Waters
  • Email : waters@crl.dec.com
  • URL : http://www.research.digital.com/CRL/people/waters/bio.html
  • Biography : Keith Waters was born in 1962 and educated in England. He received a B.A. Hons degree in Graphic Design in 1985 and a Ph.D in Computer Graphics from Middlesex University, London in 1988 for his thesis on the Computer Synthesis of Expressive Three-Dimensional Facial Character Animation. During the period 1984-88 he held summer positions at the BBC electronic graphics current affairs department and Cal Video Graphics. In addition he consulted with BenchMark Technologies and IO Research. From 1988-91, he was a member of the technical staff at Schlumberger Laboratory for Computer Science Austin Texas, where he worked on 3D visualization of seismic and borehole data. In 1991 he became a member of the research staff at Digital Equipment Corporation's Cambridge Research Lab, where he is involved in computer graphics and computer vision research. His current interests include computer vision, multimedia, computer-based facial synthesis, medical facial applications, physically-based modeling, and volume visualization. Dr. Waters has published papers in computer graphics, multimedia, computer vision, and biomedical visualization. He is a member of ACM Siggraph and IEEE, and is on the editorial board for the Journal of Visualization and Computer Animation.

Institute(s)

• Digital Equipment Corporation
  • URL : http://www.research.com

Film 1	Video QuickTime ->	(3.0 Mo)
	Jpeg Images ->	(10 Ko)

Film 2	Video QuickTime ->	(3.1 Mo)
	Jpeg Images ->	(9 Ko)

Film 3	Video QuickTime ->	(2.0 Mo)
	Jpeg Images ->	(9 Ko)

• Date : 12:12:94

• Category : Conferences

• Keys Words :

  • real-time lip-synchronization

• Theme(s) :

  • Face
  • Facial animation

Technical Information

• Bibliography :

  Imagina's Proceeding 1995, pp33-45
  
  

• Abstract :

  DECface : An Automatic Lip-Synchronization Algorithm for Synthetic Faces. DECface is a synthetic realtime (15Hz+) face that automatically synchronizes lip motions to a speech synthesizer. Based on plain ASCII text input, the algorithm creates a synthetic speech segment via DECtalk and a graphicaldisplay of an articulate mouth and face. The key component of the algorithmis the run-time facility that adaptively synchronizes the graphical display ofthe face to the audio when played. In addition to lip synchronization,synthetic muscles deform the facial geometry into canonical facial expressions. The complete process, known as DECface, is the visual compliment of DEC talk and the underlying algorithms provide a new user interfaceparadigm. Ultimately, we believe that some man-machine interfaces will mimicthe way we interact face-to-face, thereby transforming the computer from aninert box into a truly personable computer. To develop unique humanisticcharacteristics into DECface it was necessary to embed an interpreted languageto quickly prototype interfaces. With such a tool it is now possible torapidly develop facebased interfaces to mail handlers, databases, and fileeditors. In addition, more sophisticated interfaces can be rapidlyprototyped, for example a career counselor, a pseudo psychologist, or a Shakespearian joker.
  
  

  ---

  
  
  

• Some external links :

  

  http://www.research.digital.com/CRL/people/waters/bio.html