In this study, a wireless inertial sensor system was designed for application in human motion capture. A human body node displacement and attitude data decoding strategy is proposed based on the gait analysis method, synthesizing the advantages of the complementary filter and Kalman filter methods. A human body motion capture verification platform was set up, and the performance of the wireless inertia sensor network capture system was evaluated. Results show that the designed wireless inertial sensor system can accurately measure joints in the process of human movement in real-time. The proposed attitude algorithm strategy has better accuracy in precise real-time tracking of human motion.