Efficient Compressed Video Action Recognition Via Late Fusion with a Single Network.

Hayato Terao Wataru Noguchi Hiroyuki Iizuka Masahito Yamamoto

Published in: ICASSP (2023)

Keyphrases

action recognition
compressed video
late fusion
bag of words
action classification
human actions
computer vision
human detection
body parts
activity recognition
video streams
super resolution
image retrieval
similarity measure
video frames
video content
pose estimation
multiresolution
image processing