AddBiomechanics Dataset: Capturing the Physics of Human Motion at Scale
This dataset addresses a critical gap for researchers in biomechanics and computer vision by providing high-quality pose and force data for a variety of movements, enabling better estimation of human dynamics.
The researchers tackled the challenge of quantifying human motion dynamics, such as joint torques and external forces, by creating the AddBiomechanics Dataset 1.0, which includes physically accurate data from 273 subjects, over 70 hours of motion and force plate data, totaling more than 24 million frames.
While reconstructing human poses in 3D from inexpensive sensors has advanced significantly in recent years, quantifying the dynamics of human motion, including the muscle-generated joint torques and external forces, remains a challenge. Prior attempts to estimate physics from reconstructed human poses have been hampered by a lack of datasets with high-quality pose and force data for a variety of movements. We present the AddBiomechanics Dataset 1.0, which includes physically accurate human dynamics of 273 human subjects, over 70 hours of motion and force plate data, totaling more than 24 million frames. To construct this dataset, novel analytical methods were required, which are also reported here. We propose a benchmark for estimating human dynamics from motion using this dataset, and present several baseline results. The AddBiomechanics Dataset is publicly available at https://addbiomechanics.org/download_data.html.