Drivers' Motion Depth DataBase - DMDDB

DMDDB is an open-access scientific database that contains 523 depth video sequences of 27 drivers performing 236 merges and 287 lane changes. The depth data were collected in Spring 2014, in the area of Gainesville, Florida using Microsoft's Kinect sensor with the popular J4K (Java for Kinect) library. The 3D motion of the drivers was captured in the database in more than 300,000 depth frames (305,333), with 16 billion 3D points (15,739,194,425). You can browse the 3D data using our on-line DMDDB viewer.

Furthermore, it includes an easy-to-use programming interface (API) that allows you to get full access to the DMDDB database and build your own experiments in less than 10 lines of Java or JavaScript code. The DMDDB API documentation provides all the details regarding the functionality of the provided API, along with several source code examples for accessing the depth data and metadata of the database from your own Java or Javascript programs. You can read a brief tutorial on "How to write your own DMDDB experiment" in less than 10 lines of Java code, and you can also download several source code examples from the Source Code Examples website.

The interdisciplinary team that collected, segmented, and created this dataset consisted of faculty and students from the University of Florida and the University of Alabama, Birmingham listed here in alphabetical order: Angelos Barmpoutis, Shaghayegh Rostami Hosuri, Md Mozahidul Islam, Alexandra Kondyli, Virginia P. Sisiopiku, Lei Zhang, Liangke Zhao.

The DMDDB API is based on the UFDW Java Library developed by Prof. Angelos Barmpoutis, and further extended by the students and faculty who work in the SAGE program (Serious and Applied Gaming Environments) at the Digital Worlds Institute. You can find more information about how to join our graduate and undergraduate programs at this link (

Where to begin?
  • You can browse the database using our On-Line 3D Viewer.
  • You can download several source code examples from the Source Code Examples website.
  • You can also read a brief tutorial on "How to write your DMDDB experiment".
  • Optionally, you can read our Published Papers to learn more about the data collection and processing methods.

  • Click on the above image to enlarge.

    Click to browse the dataset.

    Acknowledgement: Funded in part by the US Department of Transportation / STRIDE, Award ID: DTRT12-G-UTC04 / 2013-051S, STRIDE project title "Investigating the effect of drivers' body motion on traffic safety", Principal Investigator: Angelos Barmpoutis, September 2013-April 2015.

    Disclaimer: The names JAVA and KINECT and their associated logos are trademarks of their respective copyright owners Oracle and Microsoft. None of these companies endorse, fund, or are in any way associated with the DMDDB project.

    Disclaimer: This database and accompanied software is provided for free without any warranty expressed or implied for academic, research, and strictly non commercial purposes only. By downloading this software and/or database you accept the Terms and Conditions.

    University of Florida, Digital Worlds Institute, P.O.Box 115810, 101 Norman Gym, Gainesville, FL 32611-5810, USA