The field of human movement biomechanics is shifting towards a more comprehensive understanding of human motion, incorporating physical plausibility and interaction with the environment. Researchers are exploring novel approaches to address challenges such as limited datasets and noise in wearable sensor data. Data augmentation techniques are being developed to generate more realistic and effective datasets, while mesh fitting and registration methods are being improved to enhance pose estimation and segmentation accuracy. Noteworthy papers include:

• Robust Human Registration with Body Part Segmentation on Noisy Point Clouds, which introduces a hybrid approach that incorporates body-part segmentation into the mesh fitting process, significantly outperforming prior methods.
• MotionPRO: Exploring the Role of Pressure in Human MoCap and Beyond, which constructs a large-scale human motion capture dataset with pressure, RGB, and optical sensors, and demonstrates the effectiveness of fusing pressure with RGB features for pose and trajectory estimation.