<p>With the gradual progress of technology, people have higher requirements for the goals of sports and have begun to incorporate information technology as an auxiliary tool into sports training. Coaches and athletes generally improve their athletic level by developing rigorous personalized plans, and when formulating plans, it is necessary to identify the breakthrough points of athletes during the training process. Therefore, this article developed a sports training visualization system using motion capture sensing and ultra multi-objective evolutionary algorithms. Firstly, this article understands how inertial motion capture sensors output accurate signals, analyzes the working principles of accelerometers and gyroscopes, identifies the characteristics of pulse noise inside gyroscopes, and initializes the obtained data. Then, this article analyzed the issues that should be paid attention to when optimizing objectives, and used the corresponding multi-objective evolutionary algorithm to come up with a practical solution: Strong Dominance Relationship (SDR), which can minimize the computational cost. Finally, this article designs a sports training visualization system using motion capture systems and ultra multi-objective evolutionary algorithms. This system is an auxiliary tool for modern sports training, which can effectively compensate for the problems that arise in traditional sports training processes and improve athlete performance.</p>