The cross-study of computer science and painting art is not only beneficial to the development of AI but also beneficial to the cultural innovation of painting art. The Daozi team led by Dr. Gao Feng has been exploring the intersection of technology and art. This project quantifies painting information through machine learning methods, especially deep learning technology, to analyze and explore. Specifically, from the view, clues, reading, appreciation, creation, education of six parts. "View" refers to the use of deep learning technology to design effective features for similarity painting image retrieval. "Clues," refers to an image feature that integrates edge features and local invariance features to distinguish the characteristics of painting strokes. "Reading" refers to the use of machine learning methods to design multi-task assistance and multi-scope architecture from details, parts, and wholes to distinguish between painting styles and painter characteristics. "Appreciation" refers to a painting style transfer algorithm model based on a generative adversarial network designed by combining deep learning technology and aiming at the characteristics of pen and ink in Chinese painting. "Creation" refers to the application of a painting style transfer model to assist painting creation and visual analysis and presentation. "Education" refers to the intelligent human-computer interaction method and feedback-based cognitive model for painting education. And construct and label high-quality related image data sets, and explore the interdisciplinary research of combining computer science and painting art based on artificial intelligence.