Stylized Portrait Generation and Intelligent Drawing of Portrait Rendering Robot

Abstract

In order to realize application in real time portrait rendering robot systems, a generating method of stylized portraits based on the triangle coordinate system is proposed to solve the problem of the dependence on training samples and the high complexity of the existing algorithms. The proposed algorithm automatically calculates the personalized facial characteristics of the input face and by exaggerating them, features of its stylized portrait are determined. Two triangle coordinate systems are established based on the features of the input face and the stylized portrait respectively, and the stylized portrait is generated by image morphing between the two systems. Besides, in order to improve the intelligence of the portrait rendering robot, a trajectory planning method is presented to imitate the drawing techniques of a real painter. Experimental results show that the stylized portrait generation algorithm can produce a vivid stylized portrait in less than 0.5 seconds. Finally, the portrait rendering robot imitates real painters to draw the generated stylized portrait based on the proposed trajectory planning method.