Improving videophone subjective quality using audio information
✍ Scribed by A. Vahedian; J. Arnold; M. Frater; M. Cavenor; L. Godara
- Publisher
- John Wiley and Sons
- Year
- 1999
- Tongue
- English
- Weight
- 244 KB
- Volume
- 10
- Category
- Article
- ISSN
- 0899-9457
No coin nor oath required. For personal study only.
✦ Synopsis
This article presents a new technique which uses audio information to achieve more efficient video coding for videophone and videoconferencing applications. The direction of arrival of the audio signal at an array of microphones is used to estimate the position of the speaker's lips such that the quality of the video reconstruction can be enhanced in this crucial area. Once this estimation is performed, then a two-or three-stage quantization strategy is applied to the video information which results in the compression of the subjectively more important parts, i.e., the lips and the face of a speaker, with lower distortion. Algorithms for audio source location using the speech signals received at an array of microphones form an important part of our approach. The proposed new technique is compatible with all existing video compression standards and is much easier to implement than previously proposed techniques.