EGGNOG: Elicited Giant Gallery of Naturally Occurring Gestures Dataset

The Computer Vision Group along with the Human-Computer Interaction Lab collected the EGGNOG dataset in 2016. In order to foster research in gesture recognition this dataset is being made available to other researchers.

This dataset features over 7 hours of footage over 360 videos containing natural gestures used by people when collaborating to complete a shared task. See the CSU CWC Datasetes page for a short dataset summary.

Before you decide you want this dataset, you are encouraged to download and read our paper describing the dataset:

• EGGNOG: A continuous, multi-modal data set of naturally occurring gestures with ground truth labels, Isaac Wang, Mohtadi Ben Fraj, Pradyumna Narayana, Dhruva Patil, Gururaj Mulay, Rahul Bangar, J. Ross Beveridge, Bruce A. Draper and Jaime Ruiz, 12th IEEE International Conference on Automatic Face and Gesture Recognition, May 30-June 3, 2017, Washington, DC.

Through this page you may download the EGGNOG dataset in order to support your own research. Access to the dataset is contingent to your agreeing to and following the terms set out in the EGGNOG User Agreement. Please read this agreement fully before proceeding to aquire the dataset.

One more thing, be aware the dataset is a relatively large download in the form of a singel (exact size goes here) zip file.

• Download EGGNOG.

Once you have the EGGNOG dataset, you may find the following technical document helpful.

• Gesture Annotation Key: Summer 2016, CSU CwC Project, Isaac Wang, Mohtadi Ben Fraj, Pradyumna Narayana, Dhruva Patil, Gururaj Mulay, Rahul Bangar, J. Ross Beveridge, Bruce A. Draper and Jaime Ruiz

For more information about the EGGNOG data you may contact any of us below.

• Ross Beveridge
• Bruce Draper
• Jaime Ruiz