Virtual reality (VR) technology is increasingly important in various fields, such as medicine, engineering, and science, and can be used for training and education, clinical assessment, and re- habilitation. VR is widely used in rehabilitation to evaluate and intervene in various diseases, with the potential to improve the balance and gait of patients with neurological disorders and provide additional benefits when combined with traditional rehabilitation methods. However, current VR training systems lack methods and technologies for functionally quantifying training performance. This study implemented VR head-mounted display technology and motion sensing recognition technology to achieve positioning and free movement in a virtual reality space, using motion inter- action functions in different movement modalities to achieve trunk control and balance training. Functional quantification evaluation of trunk control and balance was achieved through algorithms that include movement trajectory, trunk activity, and three-dimensional interval motion perfor- mance. The rehabilitation process of the trainees was digitally captured and updated in real-time to form a functional report, allowing medical staff and patients to understand the rehabilitation progress in real-time and promote rehabilitation effectiveness.