Alignment of nearly-repetitive contents in a video stream with manifold embedding.

Manal Al Ghamdi Yoshihiko Gotoh

Published in: ICASSP (2014)

Keyphrases

video streams
manifold embedding
head pose estimation
video data
dimension reduction
manifold learning
video frames
human motion
low dimensional
metadata
multimedia
feature extraction
high dimensional
three dimensional
video sequences
visual features
human body
image sequences