•  
  •  
 

Journal of System Simulation

Abstract

Abstract: An efficient method of auto-generating portrait cartoon based on true portrait pictures was proposed. The first step was to establish a set of portrait dictionary. As a result, the portrait was divided into six components including eyebrows, eyes, nose, mouth, face, hair, so that the material library of cartoon corresponding to each component could be established. The next was to collect some real portrait images and synthesize them using the component material, meanwhile to calculate the facial shape features and the texture features of hair corresponding to each portrait. The second step was to input a real portrait image and auto-generate the corresponding cartoon. In this method, facial components were selected by the matching of shape features, similarly the component of hair was got by the matching of texture features. The method is not only simple with more abstract and more artistic result, but also can be applied to other types of portrait style.

First Page

682

Revised Date

2014-05-13

Last Page

688

CLC

TP391.9

Recommended Citation

Sun Jingjing, Yu Jiajun, Li Bei, Xie Zhifeng, Ding Youdong. Method for Auto-generating Cartoon Based on A Set of Portrait Dictionary[J]. Journal of System Simulation, 2015, 27(4): 682-688.

Share

COinS