Topical Video Object Discovery from Key Frames by Modeling Word Co-occurrence Prior.

Gangqiang Zhao, Junsong Yuan, Gang Hua

Browse the full CVPR paper archive.