Human face stick figure generation method and device

Abstract

The invention discloses a human face stick figure generation method and device. The method comprises the following steps of: identifying each part area of a human face in a human face image; cutting out an image block corresponding to each part area from the face image; inputting the face image into a trained global portrait generation model to enable the global portrait generation model to generate a global face stick figure of the face image; for the image block of each part region, inputting the image block into a local portrait generation model corresponding to the part region, so that thelocal portrait generation model generates a local human face stick figure of the part region; and fusing the global face stick figure and each local face stick figure to obtain a final face stick figure. The invention aims to consider the difference between different parts of a human face so as to overcome the influence of image domain representation of different parts of the human face on the synthesis quality of the stick figure, and synthesize the high-quality human face stick figure with clearness, attractiveness and vivid texture in combination with the representation mode of different part regions in the stick figure.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method and device for generating human face stick figures. Background technique [0002] Converting face images into stick figures has important application value in the fields of public safety and digital entertainment. [0003] In the traditional image processing method, if the effect of the generated stick figure is to be good, the computational complexity of the image processing method is very high, which is difficult to meet the real-time requirements. With the development of machine learning technology, image processing based on machine learning Compared with traditional image processing technology, the technology has higher computing speed and higher accuracy, so many machine learning models for generating stick figures from face images have been derived. [0004] However, the robustness of the stick figures generated by these machine learning models from face images ...

AI Technical Summary

Problems solved by technology

[0003] In the traditional image processing method, if the effect of the generated stick figure is to be good, the computational complexity of the image processing method is very high, which is difficult to meet the real-time requirements. With the development of machine learning technology, image processing based on machine learning Compared with traditional image processing technology, the technology has higher computing speed and higher accuracy, so many machine learning models for generating stick figures from face images have been derived

[0004] However, the robustness of the stick figures generated by these machine learning models from face images is very poor, and the external expressions of different regions of the face in the image and stick figures are very different, and the synthesis effect is not good.

Images