In this paper, we construct a robot photographic system to search for the optimal viewpoint of a scene. Based on some known composition rules in the field of photography, we propose a novel aesthetic composition evaluation method by the use of Kullback-Leilber divergence. For viewpoint selection, we put forward a method called Compositionmap, which can estimate the aesthetic value of scenes for each candidate viewpoint around the target group. At last, the effectiveness of our robot photographic system is confirmed with practical experiments.
CITATION STYLE
Lan, K., & Sekiyama, K. (2016). Optimal viewpoint selection based on aesthetic composition evaluation using Kullback-Leibler divergence. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9834 LNCS, pp. 433–443). Springer Verlag. https://doi.org/10.1007/978-3-319-43506-0_38
Mendeley helps you to discover research relevant for your work.