DCT-based video features for audio-visual speech recognition.

Martin Heckmann Kristian Kroschel Christophe Savariaux Frédéric Berthommier

Published in: INTERSPEECH (2002)

Keyphrases

video sequences
audio visual speech recognition
low level
image features
video content
feature space
feature vectors
feature set
video data
video streams
human activities
feature extraction
multiresolution
space time
semantic information
key frames
visual data