National vocal music is a traditional Chinese music genre and has a great influence in China. However, the current teaching of national vocal music is still conducted in the traditional small-class teacher-apprentice relationship, which hinders the sustainable development of national vocal music. Therefore, this paper aimed to use computer vision technology to build a national vocal music learning resource library and design a personalized teaching system for learning resources. For computer vision technology, this paper used image segmentation and video tracking technology to design an image recognition algorithm for hand and face, which provided help for students in the classroom. For the design of personalized teaching system, this paper introduced the database of ethnic vocal music learning resources and the steps of using vocal music learning resources in detail, and designed the B/S structure of the system in detail. From the test results of the system in this paper, it can be seen that the test success rate of the five functional modules of the system was above 99%, and the response time and delay of the system were within the acceptable range, which showed that the system designed in this paper could provide certain help for the teaching of national vocal music.