This paper focuses on the platform construction in the “three stations” fusion environment, and explores the effective design method of virtual learning environment and intelligent interaction module in the online teaching platform of aesthetic education in colleges and universities. The immersive aesthetic education learning scenario is constructed through big data, cloud computing and other three-platform integration technologies, and the virtual learning environment is generated by combining the learning resource recommendation algorithm based on the improved cosine similarity metric. Based on the Kinect device, the intelligent interaction module in the online learning platform is designed, focusing on the construction of virtualized body posture interaction and voice interaction in the interaction module to enhance the immersion and experience of the interaction design. The experimental results show that the learning resource recommendation algorithm constructed in this paper has higher recommendation accuracy compared with other algorithms, and the pose recognition algorithm can achieve more than 90% recognition accuracy in all subjects. Meanwhile, speech text recognition and speech emotion recognition also show good recognition performance. The virtual learning environment and intelligent interaction module in this paper can significantly improve the teaching level of aesthetic education in colleges and universities by applying them to the online platform for teaching aesthetic education in colleges and universities.