科研之友
微信
新浪微博
Facebook
分享链接
摘要
Traditional cartoon animation painting has always been a tedious job. In order to improve the efficiency of the process, the development of an automatic cartoon generation system including automatic inbetweening and coloring is required. Automatic matching of cartoon characters in key frames is the prerequisite for the system. This paper provides a novel matching algorithm with iterative maximum a posteriori (MAP) estimation and the maximum likelihood (ML) estimation. Specifically, this algorithm formulate cartoon matching as a many-to-many labeling problem. To refine the results of matching, an optimization approach is adopted to alternatively conduct the MAP estimation and the ML estimation. Besides, we construct the correspondence by using the local shape descriptor, and the rotation and scale invariance in matching can be achieved. The experimental results on real-world datasets demonstrate the effectiveness of the proposed methods for automatic cartoon matching.
