Reaching new heights
How Project Skate is helping to change the face of skateboarding with machine learning
3-minute read
"I don’t think people know how good the girls are."
Sky Brown, pro skater
Interest in skateboarding continues to soar after the sport saw its debut in the Tokyo Olympic Games in 2021. But despite growth, equal participation across genders remains skewed. And this trend is reflected in limited opportunities to secure brand sponsorships for female professional athletes.
Skating in competitions since the age of eight, Sky Brown won a bronze medal at the 2021 Tokyo Olympics, representing Great Britain
However, the root of this imbalance isn’t a lack of talent. Whilst athletes like pro skater Sky Brown are advocating for more girls to enter sports, their efforts are hindered by the misconception that they are not as skilled as their male counterparts.
Skateboarding is a combination of athleticism, unpredictable body movements and personal flair. While style will always remain a matter of preference and opinion, technology can play a role in capturing performance data that the human eye cannot measure. This data can also be used to help demonstrate the true ability of female athletes and address misconceptions about athletic ability based on gender.
Measure Awesome
Project Skate is a machine learning tool that can measure the speed, height and rotation of a skateboarder, and identify grabs and tricks in real-time. Developed by Googlers in Tokyo and Sydney, the tool’s functionality was refined with local skateboarders. As a world-class athlete and passionate advocate for girls in skateboarding, Sky Brown was one of the first to test the tool.
Project Skate captures Sky's body position at every part of the trick, generating the data behind her movements
“People don’t really look at height right now. They just look at difficulty. This is going to push me.”
Sky Brown, pro skater
Using Google's machine learning tools and a combination of cameras, Project Skate determines the position of a skater's body in 3D space.
The technology utilises Tensorflow, an open source platform to build, train and deploy machine learning models, and Blazepose, a Tensorflow tool that can understand the stance of a person, and draw points on them.
“Using machine learning, computers can be trained to recognise skateboarders and their body movements. This data allows us to capture the exact measurements behind their performance" says David Arcus, Creative Technologist for Google’s Brand Studio APAC.
Three cameras were installed to ensure a view of the skateboarder from all angles
Using Tensorflow and Blazepose, Project Skate identifies the skateboarder in frame and draws points on parts of their body
Three cameras were installed to ensure a view of the skateboarder from all angles
Using Tensorflow and Blazepose, Project Skate identifies the skateboarder in frame and draws points on parts of their body
The movement of their shoulders gives information on rotation, the position of their hands provides data on grabs, and the placement and angle of their feet is used to understand the board’s position.
The technology combines these elements to identify the trick performed, and display exact data behind it, like speed, height above the top of the ramp, and airtime.
Project Skate was able to identify Sky's signature tricks, and provide data behind them that she had never seen before.
"I want to inspire girls to get into skateboarding but also other sports. I think we can do anything boys can do."
Sky Brown, pro skater
"Project Skate enables skateboarders – who may be underestimated due to human bias – to demonstrate the extent of their skills with certainty."
David Arcus - Creative Technologist, Brand Studio APAC
The sky's the limit
Project Skate has demonstrated how empowering female skateboarders with the stats behind their skills, motivates them to push even further. But beyond personal development, these unseen metrics have the opportunity to change perceptions on the true capabilities of all skateboarders, regardless of gender or ability.
“We’re excited to continue developing the technology and see how it can help showcase the performance of athletes, and bring equal participation across the board.”
Tara McKenty - Creative Group Lead, Brand Studio APAC
The tool has the potential to aid judges and commentators in competitions on metrics like height, rotation and trick classification. And the analysis of these metrics and complex body movements could be adapted for use in other sports. By providing a new perspective with highly accurate data, the technology has the power to help level the playing field for athletes, improving representation for all.
