In this project we will exploit computer vision and multimedia solutions to improve Passenger Experience. Our effort in the PITAGORA project is aimed at the development of three novel applications.
A real-time people counting system for crowded environments
Our solution is based on depth cameras and it relies on efficient computer vision algorithms to automatically count people in crowded environments. The approach we propose is a counting-by-detection method on depth imagery that is designed to be deployed as an autonomous appliance in a crowd analysis scenario like the one commonly found in an airport. Our system performed foreground/background segmentation on depth image streams in order to coarsely segment persons, then depth information is used to localize head candidate regions which are then tracked in time on an automatically estimated ground-plane. The system runs in real time, at a frame-rate of about 20 fps.
An innovative baggage size estimation system
This system allows a passenger who will access an interactive kiosk to automatically estimate the volume of his own handbaggage. Our solution uses advanced computer vision techniques, a depth camera and it runs in real-time. The application exploits a cooperative paradigm between the user and the system. The interface will guide the user in the acquisition of depth images to estimate a bounding box in which the baggage will fit in and then the system will compare it with the baggage allowance for her/his flight.
A dedicated Socialnet for passengers
This module will facilitate bi-directional channels of communication between passengers and the airport through the use of social media such as Facebook or Linkedin. This social networking infrastructure will enrich the passenger experience providing personalized information in order to optimize the time spent by passengers in the airport. The user will be able to interact with the PITAGORA social network through a web application using his personal device (e.g. a notebook or a smartphone). The social network will access airport facility information such as location and times of shops, flight boarding time and gate in order to plan the most pleasant passenger stay. The application will employ user profiling as a mean to offer products, places and events dating.