Robot portrait rendering based on multi-features fusion method inspired by human painting

摘要

Traditionally, image drawing robot draws a portrait only by generated image, without concerning the certain order of people's facial features. Therefore, to make a portrait, a robot drawing model was proposed. Through geometric and textural features generating algorithm, lines to represent the geometric contours and texture features of human face are captured. Besides, a fusion method is proposed to better strengthen the feature integrity of partial region. The lines represent geometric features and textural features are fused so that a face sketch which is more suitable for robot rendering is obtained. Experimental results show that with our rendering model, robot draws a portrait more effectively. Comparing with the previous robot drawing method, the drawn image contains more information of facial features.