QueueHop: Computer Vision Wait Time Estimation for Clark Hall Pub

Thumbnail Image
Chase, Alex
Macduffee, Bryn
Garmaise, Jonah
Everitt, Julia
Clark, Griffin
engineering , computer vision
Computer vision is a rapidly growing field that involves the use of machine learning algorithms to analyze visual data from images and videos. One application of computer vision is in crowd counting, which involves using algorithms to automatically estimate the number of people in a given area. In this project, we sought to use computer vision to determine the length of the line at Clark Hall Pub and estimate the wait time for individuals in line. Additionally, time of flight sensors were used to automatically count the pub's capacity. These metrics were displayed to users via a mobile application built using JavaScript and React Native, providing them with information to help them decide about their plans for Friday afternoon. The machine learning model achieved a 7.2% mean average error, the sensor capacity device achieved a 78% accuracy, and the mobile application achieved a frame rate of 60 frames per second. This provided the end user with a user-friendly experience and the ability to display live data. In future work, full integration of these subsystems is an area of focus.
External DOI