|Salil | Current Research | Publications|
|Salil Deena's Research Page|
Current ResearchI am currently working on visual speech synthesis or facial animation driven by speech, which involves driving a talking head from recorded audio after having learnt a mapping between audio and visual parameters using Machine Learning techniques. I am investigating novel techniques for audio-visual mapping using joint probabilistic models of audio and video.
Visual Speech Synthesis Method
Visual Speech Synthesis Experiment (only for demonstration)
Examples of real and synthetic videos
| Mr. Salil. Deena |
School of Computer Science | The University of Manchester | Room 2.100 | Kilburn Building | M13 9PL | sdeena AT cs DOT man DOT ac DOT uk |