Researchers at the Pittsburgh campus of Disney Research are using computer vision to analyze the patterns of field hockey players, in hopes of creating a new way for coachers and commentators to make sense of game data in real time. Furthermore, this technology can be used not only in field hockey, but any other kind of team sport with continuous play.
With a focus on player roles, the research zeroes in on the tactics, strategy, and style for players and their teams. Eight cameras recording high-definition video are used to record matches and data from these is analyzed against other matches. The compiled information can give insight into strengths and weaknesses of teams and solid strategies for how to better face their opponents.
While there is a lot of talk about the ways computer vision can save lives, in some instances, it is already doing just that.
Last month, a computer vision drowning detection system, known as Poseidon, saved a man in Australia from drowning after an epileptic seizure caused him to sink to the bottom of a pool.
And he wasn’t the first, either. In total, 25 people were saved as a result of this system being implemented in pools.
Currently, Poseidon is in more than 220 pools across America, Europe, Japan, and now Australia.
According to the company, “the Poseidon system is based on a network of overhead or underwater cameras connected to a computer equipped with the Poseidon software [that] analyzes the trajectories of the swimmers and sends an alert to lifegards when a swimmer is in trouble.”
In light of this summer’s Olympics, currently being held in London, it seems appropriate to touch on the ways in which computer vision has contributed to the international sports competition. Most recently, it has been related to fitout, which, for kayakers, mean the building of custom parts of the kayak that fit to the bodies of the competitors.
Researchers at the Australian Institute of Sport have been doing just that, working to make the athlete, the kayak, and the paddle all act as one cohesive unit.
According to Ami Drory, a biomechanist working on this project: “A good fitout allows the athlete to use their full range of motion while transferring as much force as possible into the water.”
Unfortunately, working on the fitout requires a lot of time and a lot of wasted material, which is why the institute decided to call upon a specialist at Canadian 3D-scanning developer Creaform. Together, they scanned athletes bodies in position as well as the kayaks they planned to compete in, to make the best possible fit.
And as it turns out, one of the athletes scanned, 18-year-old Jessica Fox, went on to take the silver medal this week. That’s not to say that she wouldn’t have done well without it, but for all anyone knows, this fitting could have propelled her from merely participating in the Olympics to being a medal holder.
Facial recognition software created a stir in the wake of the London riots this past summer, as it provided authorities with a way to identify perpetrators in a crowd. And although the technology hasn’t evolved enough to regularly outdo traditional methods of identification, it’s only a matter of time before it might.
This brings up the question of how riots, which are spontaneous and unpredictable by nature, might somehow evolve if individuals on the streets are aware that they’re being videotaped and watched nearly everywhere they go.
And while some footage might not reveal much, advances in the industry have made it so there is the possibility that multiple cameras could collaborate on information, tracking the movements of an individual across a specified amount of distance and time. This means that a rioter who has disguised himself on the streets could – in theory – be followed home, where different or better footage could more easily identify him.
Someone bent on not being recognized could still evade detection via this method, but facial recognition technology has opened up other possibilities. One of the scariest is the ability for programs to match photographs with personal information compiled from the Internet, something which many people disperse freely without thinking of how that information can potentially be used.
But the outlook isn’t entirely grim, as these same computer algorithms can be used to find missing persons in a crowd or even criminals on the run. Still, those are extreme cases. Even on a smaller scale, facial recognition could be used to track athletes on the field or in a race. And that isn’t all. What practical or positive uses for facial recognition can you think of?
Image courtesy of Ecole Polytechnique Fédérale de Lausanne
In addition to tracking athletes at play, the technology can also be extended to use by artificial sportscasters, who would relate play-by-play updates based upon the information obtained by these cameras. Meanwhile, outside of the sports word, this could also be used in the fields of marketing research, as well as more specifically in observing and identifying pedestrians in traffic.