Actor-centric Short-clip Generation System

This paper proposes a system which recognizes faces in videos and automatically generates an actor-centric short clip. First, deep learning based face recognition is performed with movies and dramas, which recognizes 1,000 Korean actors. Secondly, temporal unit segments about actors’ appearance time in a video are made. Next neighboring unit segments in which the same actors appeared are merged. Finally, each actor’s shortclips are generated based on the merged segment information.

[1]  Yong-Seok Seo,et al.  Recognizing Faces using CNNs , 2016 .

[2]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Josephine Sullivan,et al.  One millisecond face alignment with an ensemble of regression trees , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.