Quentin Galvane, Rémi Ronfard, Christophe Lino, Marc Christie
AAAI Conference on Artificial Intelligence, 2015
We describe an optimization-based approach for auto- matically creating well-edited movies from a 3D an- imation. While previous work has mostly focused on the problem of placing cameras to produce nice-looking views of the action, the problem of cutting and past- ing shots from all available cameras has never been ad- dressed extensively. In this paper, we review the main causes of editing errors in literature and propose an edit- ing model relying on a minimization of such errors. We make a plausible semi-Markov assumption, resulting in a dynamic programming solution which is computation- ally efficient. We also show that our method can gen- erate movies with different editing rhythms and vali- date the results through a user study. Combined with state-of-the-art cinematography, our approach therefore promises to significantly extend the expressiveness and naturalness of virtual movie-making.