Fujitsu’s Pursuit of Fairness: Measuring the Mechanics of Gymnastic Perfection

Fujitsu 3D Gymnastic Modeling

The robots aren’t quite taking over by 2020. But they will be assisting gymnastic judges at the 2020 Olympics.

The International Gymnastics Federation (FIG) intends to employ laser technology in order to provide data and recommendations to judges instantaneously to supplement their own judgment based on what they see with their own eyes.

Fujitsu, which is a Tokyo 2020 Gold Partner, has been focusing its 3D sensory technology on the world of gymnastics in order to provide real-time feedback on techniques a gymnast is attempting and making, the elements of which can be hard to catch by the naked eye. This technology, an advance on the more familiar technology of applying motion capture balls all over the body of an athlete, does not require the subject to apply anything. Instead a device that emits lasers from a single point, one that looks like a camera, feeds data that processes, essentially, 360 degree views of the subject.

With that amount of data captured, the software is able to assess exactly how the body has moved through the air. Benchmarking against standards of excellence inputted into the software, the system is then able to provide a report to judges what techniques were attempted or made, and then assess points to those series of techniques. In essence the 3D sensory system can do the gymnastic judge’s job.

Clearly, there is an advantage a non-human system has over a human – a “robot” will not get tired or cranky at the end of the day, nor will it allow unconscious biases about nationality, race, appearance, etc to seep into its judgment. But there has been some concern, as there should be.

According to this article, whenever you place your measuring and evaluating systems in the hands of algorithms, you are subject to hacking of some sort.

“In gymnastics, you can have 10 to 100 independent moves the system is trying to score. If the algorithm were manipulated by even a small portion you could affect the overall outcome score and it would be very hard to detect,” said Betsy Cooper, the executive director for long-term cybersecurity at UC Berkeley.

Fujitsu 3D Sensing Technology for Gymnastics
Fujitsu 3D Sensing Technology for Gymnastics

Any technology that relays information from an external source – like this 3D sensor does – to a computer is at risk. Mix the technology with a scoring panel of judges, and there is room for manipulation. “You can manipulate the algorithms to change the score one out of every five times, making it hard to detect. That area is most disconcerting. Whoever has an interest in the outcome of these major sporting events will also have an interest in trying to take advantage of any such system,” Cooper said.

This is why, the plan is to have 3D sensory system only assist judges, not replace them.

Perhaps the greatest impact will be in faster development of gymnasts. Coaches and gymnasts can examine the data, understand the micro-movements that keep their points down, and apply their practices to improving their movements to get their points up.

In other words, the next generation of gymnasts will grow up on this feedback, understanding what specific things they need to do for perfection at very early stages in their career. In this new age of digital analytics and 3D modeling, athletes are able to approach perfection at a faster rate than ever before.