Unsupervised Semantic Parsing of Video Collections. ICCV, 2015.

Unsupervised Semantic Parsing of Video Collections. ICCV, 2015.

http://openaccess.thecvf.com/content_iccv_2015/papers/Sener_Unsupervised_Semantic_Parsing_ICCV_2015_paper.pdf

Human behavior has an underlying structure. In this paper, we propose a supervised learning method for parsing a video into a semantic “storyline” composed of objective steps.

We evaluate this method on a large number of complex YouTube videos and show results of unprecedented quality for this intricate and impactful problem.