Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | SS-2.5 |
| Session: | Multi-Sensory Processing for Context-Aware Computing |
| Time: | Tuesday, May 18, 14:20 - 14:40 |
| Presentation: |
Special Session Lecture |
| Topic: |
Special Sessions: Multi-sensory Processing for Context-Aware Computing |
| Title: |
CHARACTERIZATION AND EXTRACTION OF MOUTH OPENING PARAMETERS AVAILABLE FOR AUDIOVISUAL SPEECH ENHANCEMENT |
| Authors: |
Frédéric Berthommier; ICP | | |
| Abstract: |
The strong association existing between subbands audio envelope parameters and video parameters extracted using the full DCT (Discrete Cosinus Transform) can be exploited for audiovisual speech enhancement, thanks to a good prediction of amplitude variati ons by a statistical linear model. Since the video parameter space is highly multidimensional, the causality of this association must be clarified. At first, a new method of retro-marking is proposed in order to build a transformation function of DCT par a meters into explicit classical ABS mouth opening parameters. Secondly a reduction to single parameter spaces is performed by selection of the best parameters. We show in two noisy conditions that the degradation of the enhancement performance due to the t ransformation and to the reduction is moderate. 1ˇ |
| |
| Back | |