Art Portrait Design Based on Mobile Internet in Digital Media

Abstract

With the improvement of people’s living standards and art appreciation, more and more people are engaged in the design and collection of art portrait. At the same time, the development of new generation information technology, such as digital media and mobile internet, has also made art portrait design techniques more diverse and more rich artistic styles. However, facing a huge number of art portrait, the current management work is mainly carried out manually by professionals, with high people and financial costs. Therefore, it is of great significance to study the effective and correct design and management of art portrait and apply digital mobile technology to provide users with more accurate art portrait. For the demand of art portrait design, this paper takes print, oil painting, ink painting and watercolor painting as the research object, and proposes a deep learning model Mask R-CNN. Based on the multi-disciplinary fields such as deep neural network, digital media and art, this paper analyzes and studies an art portrait design system based on mobile internet. In the paper, the in-depth learning model Mask R-CNN is applied in the system. By comparing the accuracy of the training of Mask R-CNN and U-Net models, this paper analyzes the effectiveness of the two models in extracting art portrait features. In addition, Mask R-CNN model is applied to the art portrait design system based on mobile internet. The final experimental results show that the art portrait design system using Mask R-CNN model has higher prediction and detection accuracy and has better practicability for art portrait design and art communication.