In which biomechanical analysis software can human motion files collected using markerless motion capture technology be simulated and modeled?

With the rapid development of computer vision and deep learning technologies, markerless motion capture ( Markerless Motion Capture ) technology is gradually becoming an important tool in biomechanics research. Compared to traditional marker-based motion capture systems, the markerless approach does not require reflective markers to be attached to the subject's body, greatly improving the convenience and naturalness of data collection. It is especially suitable for practical applications such as sports training, rehabilitation assessment, and sports biomechanics analysis.

Markerless motion capture systems use synchronized multi-camera shooting combined with 2D or 3D pose estimation algorithms (such as OpenPose 、 HRNet 、 AlphaPose etc.), to extract human keypoint information and reconstruct three-dimensional motion trajectories. These data are usually output in standard formats (such as C3D、 TRC 、 CSV etc.) and can be compatible with various biomechanics analysis software for subsequent dynamics, kinematics calculations and simulation modeling.

Currently, the widely used biomechanics analysis software internationally mainly includes:

- OpenSim An open-source simulation modeling tool that supports analyses such as inverse kinematics and inverse dynamics, and can seamlessly integrate with markerless motion capture data.

- AnyBody Modeling System Based on inverse dynamics principles, it is widely used for predicting human muscle forces and load analysis.

- BOB Human Biomechanics Analysis Software A professional human motion biomechanics analysis tool from the UK, supporting motion data visualization, joint moment calculation, and other functions. It is suitable for sports science, clinical rehabilitation, and modern intelligent manufacturing human-machine interaction fields.

In recent years, markerless motion capture technology has made significant progress in both academia and industry. For example, Needham and others proposed a markerless motion capture process based on multi-view cameras in 2022 and validated its consistency with traditional motion capture systems in measuring mechanical work during tennis motion analysis ( Journal of Sports Sciences, 2025 ). Cronin and others proposed a markerless motion capture process based on multi-view cameras in 2024 also reported lower limb joint angle measurement results based on markerless systems in Sports Biomechanics, 2024 ). Besides the research field, this technology is gradually being applied in professional sports. For example, some professional tennis teams have started to use markerless systems to monitor athletes' training loads and movement efficiency to avoid overtraining and injury risks.

Of course, in current practical applications, markerless motion capture systems still have certain errors when dealing with high-speed movements, occlusions, and fine upper limb actions. Especially when calculating internal mechanical work (such as energy generated by relative limb movements), its random error is relatively large ( LoA up to ± 21% ), so caution is still needed in some high-precision applications.

Overall, the motion data output by markerless motion capture can already be analyzed and simulated in most mainstream biomechanics analysis software, significantly lowering the usage threshold and promoting the development trend of sports science research "from the laboratory to the field."