Semanticbox: Bounding Box-Guided Caption Enhanced Action Recognition for Instructional Videos.

Jonathan McGee, Chongyu He, Peter A. Youngs, Scott T. Acton, Matthew Korban

Browse the full ICIP paper archive.