In a lip-reading system, one key issue is how to extract the visual features, which greatly impact on the lip-reading recognition accuracy and efficiency. In this paper, we propos...
The multimodal nature of speech is often ignored in human-computer interaction, but lip deformations and other body motion, such as those of the head, convey additional information...
Iain Matthews, Timothy F. Cootes, J. Andrew Bangha...
The major problem in building a good lipreading system is to extract effective visual features from enormous quantity of video sequences data. For appearance-based feature analysi...
Yun Fu, Xi Zhou, Ming Liu, Mark Hasegawa-Johnson, ...
Automatic lipreading is automatic speech recognition that uses only visual information. The relevant data in a video signal is isolated and features are extracted from it. From a s...
A practical lipreading system can be considered either as subject dependent (SD) or subject-independent (SI). An SD system is user-specific, i.e., customized for some particular ...