GlobalFestival Dataset
This data set is composed of pedestrian tracking analysis of video data, touch logs, and qualitative analysis logs.  All of this data has been derived from 21 hours of video data and on-screen logging of events during deployment.

The raw video data cannot be made publicly available due to privacy issues.  The videos were taken from roughly 9 meters above the display using a CCTV style camera.

Pedestrian Tracking Data
Automatic video analysis was completed for each of the four days of the deployment.  For each day, there is a file containing the pedestrian data (.csv) as well as the parameters that were used to complete the analysis (.log).  All of this data was generated and visualised using the tools available at: 

https://github.com/julierthanjulie/PedestrianTracking

The CSV files of pedestrian tracking have the following format:
Pedestrian ID; Frame Number; X position; Y position;

These logs represent the raw output of the pedestrian tracking before secondary validation was completed by the pedestrian tracking visualization tool on GitHub.

Touch Logs
All of the touch logs for each condition are given in the files.  These touch logs are generated from TUIO input and given in XML format.

id - ID number of the touch point
phase - phase of touch point (see TUIO prototcol for details)
x - Latitude of touchpoint as a number between 0 and 1
y - Longitude of the touchpoint as a number between 0 and 1

Video Tags
A qualitative analysis has been completed for all videos.  The results of this analysis is given in a file for each video.  

The CSV files has the following format:

id; tag code; tag description; starting frame; ending frame or note for events without end time; video file name;