The 2022 NFL Season kicked off this week with the Bills and Rams facing off from Los Angeles’s state-of-the-art SoFi Stadium, which, as we’ve previously discussed, is one of the most technologically advanced arenas in pro sports. From digital-twin technology to a video board that’s been described as the “eighth wonder of the sports world,” SoFi’s bells and whistles are among the great examples of the NFL’s ambitions to leverage AI, machine learning, and computer vision to improve and enhance gameday for spectators. 

A shot of SoFi Stadium

But the league’s technological goals go far beyond just the fan experience. Over the past few years, the NFL has been aggressively pursuing new AI models that can be deployed to proactively improve player safety—and even reimagine how coaches strategize.

The league has partnered with Amazon Web Services (AWS) to create the Digital Athlete, an AI model that replicates any NFL player within a virtual environment that’s chock-full of real-world scenarios. Within this digital setting, league doctors and coaches can test out new techniques, equipment, and even the implications of new rules before they play out on the field. 

The tool marries image and video data collected from live broadcasts of NFL games with data sourced from sensors embedded within helmets, mouth guards, and shoulder pads worn by players on the field. Algorithms can run nearly endless simulations to understand the impact on player health and safety, while opening the door for greater collaboration league-wide to make the game safer for players. 

“Having the computers understand how many times a player hits his helmet during the course of a game [helps] find ways to reduce the amount of helmet contact,” Jeff Miller, NFL executive vice president, told New Scientist. With this data, the NFL can make plans to prevent head trauma that has been linked to chronic traumatic encephalopathy (CTE), a neurodegenerative disease that’s suspected to be endemic among NFL players.

A graphic showing how computer vision can help analyze NFL gameplay to promote player safety

Using computer vision for a safer NFL

The Digital Athlete was actually born from the results of a contest held at the end of 2021 in partnership with the NFL and AWS, which specifically challenged data scientists to create detection models for head injuries using game footage. Because the NFL regularly reviews videos and images of all major injuries to record at least 150 different variables per incident, there was a wealth of visual data for the more than 1,000 contest participants to use in developing their models.

The winning solution was able to conduct a comprehensive and accurate analysis of head injuries 83 times faster than a person alone, according to the results of the findings. Kippei Matsuda from Osaka, Japan took home the top prize of $50,000 for their work. 

“The innovative ideas brought to this competition from data scientists around the world will be transformative, driving a staggering improvement in accuracy of computer vision models over just a three-month competition,” said Jennifer Langton, SVP, NFL Player Health and Innovation. “The success of this challenge speaks to the power of the crowdsourcing model that the NFL has deployed over the last decade to drive innovation in player health and safety and we are thrilled to have had some of the brightest minds in data science from around the world working on our challenge.”

Improving health outcomes off the field

While the Digital Athlete will undoubtedly unlock insights around the impact of athletes colliding on the field, the potential for computer vision solutions to improve public health outcomes is just as robust. 

Many of the same techniques deployed at SoFi to understand stadium occupancy and flow for an improved fan experience can be leveraged to proactively identify security threats, for instance, or confirm folks are adhering to PPE protocols within a specific facility. 

Video cameras combined with Plainsight vision AI models provide unblinking, highly accurate and definitive counts for occupancy verification, allowing customers to verify whether an individual is safe to enter a facility and whether their behavior adheres to guidelines. These are just a taste of the custom solutions that computer vision can unlock that have positive outcomes across industries and settings. 

Plainsight supports enterprises looking to optimize and ensure the safety of facilities and staff with custom-built models, expert professional services, and an emphasis on responsible AI. Schedule a conversation with our team to learn more about some of our customer engagements and discuss how we can help you achieve your organization’s computer vision objectives.

 View All Blogs