Talking head video generation aims to animate a human face in a still im...
Predominant techniques on talking head generation largely depend on 2D
i...
Weakly-supervised action localization aims to localize and classify acti...
Talking head video generation aims to produce a synthetic human face vid...
Weakly supervised temporal action localization (WS-TAL) is a challenging...
Weakly supervised video anomaly detection (WS-VAD) is to distinguish
ano...
The objective of action quality assessment is to score sports videos.
Ho...
We address the weakly supervised video highlight detection problem for
l...
Important people detection is to automatically detect the individuals wh...
Humans can easily recognize the importance of people in social event ima...