drivelineresearch/openbiomechanics

The open source initiative for anonymized, elite-level athletic motion capture data. Run by Driveline Baseball.

What it solves

It provides a standardized, public research dataset for baseball biomechanics, filling the gap for high-quality, cleaned motion-capture and physical assessment data for pitching and hitting analysis.

How it works

The project distributes a collection of C3D motion-capture files, processed time-series data (kinematics and kinetics), and point-of-interest (POI) metrics. It includes a Python loader (obp) and a set of scripts to help researchers download, join, and analyze the data. The dataset is organized into modules for pitching, hitting, high-performance assessments (force-plate metrics), and computer vision calibration.

Who it’s for

Researchers, data scientists, and biomechanists interested in athletic performance, human movement, and sports science, specifically within the context of baseball.

Highlights

  • Comprehensive Data: Includes 411 pitching trials and 677 hitting trials from nearly 100 athletes, plus nearly 2,000 high-performance assessments.
  • Multi-modal Inputs: Combines raw marker trajectories (C3D), processed signals, and force-plate data.
  • Tooling Included: Provides a dedicated Python helper for data loading and runnable examples for common analysis workflows.
  • CV Integration: Includes introductory OpenCV examples and multicamera calibration work for computer vision applications.

Related

  • Project
    tudelft-iv/view-of-delft-dataset

    A multi-sensor automotive dataset containing synchronized LiDAR, camera, and 3+1D radar data with 3D bounding box annotations for road user detection in urban traffic.

  • Project
    liangnjupt/VisTouch

    A large-scale synchronized dataset of video, audio, and tactile force data from robotic sliding contact used for material recognition and cross-modal learning.

  • Project
    facebookresearch/momentum

    A library of foundational algorithms and numerical optimization solvers for human kinematic motion and body tracking.

  • Project
    ShiqiYu/OpenGait

    A flexible and extensible gait analysis framework for human identification and health screening, supporting multiple modalities like silhouettes, skeletons, and LiDAR point clouds.

  • Dispatch
    LeRobot Community Datasets: Building the ImageNet of Robotics

    Hugging Face is fostering a community-driven effort to create a diverse, open-source repository of robotics datasets via LeRobot to solve the generalization challenge in robotic policies.