While I was in Paris for the Olympic Games Paris 2024, I saw an interesting demonstration of how organizers determine the number of people in different venues and even parts of venues. This is important for tasks like determining how many sandwiches are needed at concession stands and managing crowds at entry and exit points.
As a quick note, we took a bit more footage than we used in our main video, which you can find here:
We thought it was worth sharing. We have to say that all of these are sponsored by Intel since Intel sponsored our trip.
Traffic Analytics at the Olympic Games Paris 2024 Venues
In the video above, we filmed at Stade de France. “We” means the Olympic Broadcasting Services because I could not film in ticked venues. I learned a lot during this entire process, to say the least. Still, if you hear the roar of a crowd in the background of our run-through of the Intel AI Platform Experience at Stade de France, there is a good reason. The men’s Rugby Sevens was taking place at the same time.
Since there are a large number of people at these events, both as spectators but also with athletes, coaches, media, staff, and more, it is useful to know simply how many and where people are. While I could not take photos of the cameras inside the venue, there were a ton of cameras that were newly installed. For example, there was a nice seven-camera installation on one of the pathways leading to Stade de France (the ramp is in the distance.)
After going, I learned that there was a Traffic Analytics platform running. Here we could see the number of visitors in Stade de France, how long their average stay, and even some specific areas with additional detail such as working rooms for those working at the Olympic Games.
For example, we can see that the Photo Workroom had 666 visitors, but there were currently 5 visitors inside with an average stay of 7 minutes.
These dashboards were then used by organizers to determine things like when to restock concessions.
Final Words
Earlier this year we did another piece looking Inside a Secret Stadium Data Center and showed how that stadium also used cameras to do traffic analytics. Perhaps the key takeaway in all of this is that using computer vision AI to estimate crowds in stadiums is becoming commonplace. Instead of just knowing there are a few or many people in a stadium, logistics teams can see how many people are inside and where they spend their time.
If you want to learn more, we have a short for this as well:
