University of Waterloo, Orioles partner on AI project to track pitcher biomechanics
Staying competitive in the American League East means staying on the cutting edge of technology.
That's why the Baltimore Orioles have partnered with researchers at the University of Waterloo to develop artificial intelligence technology, called PitcherNet, that will help the Major League Baseball team track the biomechanics of pitchers at every level of the organization, from the collegiate ranks and through its minor-league system.
PitcherNet takes the main camera feeds fans are used to seeing — either from behind the pitcher or behind the batter — and combines it with video footage shot on a smartphone by a scout or team official. Using AI, PitcherNet then extrapolates the pitcher's entire throwing motion, recreating the level of detail that the 12-camera Hawk-Eye system is capable of producing. University of Waterloo researchers John Zelek, right, and Jerrin Bright pose for an undated handout photo in front of an onscreen example of the software they engineered to analyze baseball pitchers. To help train AI algorithms at the heart of the technology, researchers created three-dimensional avatars of pitchers so their movements could be viewed from numerous vantage points. THE CANADIAN PRESS/HO-University of Waterloo, *MANDATORY CREDIT*
'What we do is we try to extract 2D information, 2D joint position information, and then extrapolate using 3D avatars,' said Waterloo PhD student Jerrin Bright. 'The 3D avatars are like blobs encompassing a human itself and we use those blobs to basically find the 3D human pose, which can be represented with the shape of the person, so that's the idea behind it.'
Sig Mejdal, the Orioles assistant general manager, said that in his experience it's a necessity to evolve, even in a sport as old as baseball.
'It's mandatory. It's completely mandatory,' said Mejdal, who worked for NASA and Lockheed Martin's satellite operations unit at Onizuka Air Force Station in Sunnyvale, Calif., before moving to the St. Louis Cardinals and then the Houston Astros. 'The success that I've been a part of in St Louis, Houston and Baltimore, so much of it has come from innovation and change.
'We're the Baltimore Orioles, we're not a large market team, and we're in the American League East. We have to do everything right, and it would be foolish not to explore every avenue of improvement.'
As part of that need to stay ahead of the competition, Mejdal wanted to expand the Orioles' ability to track the biomechanics of their pitchers.
Currently, every MLB ballpark and over 60 minor-league stadiums are equipped with Hawk-Eye technology, a series of cameras strategically located throughout a venue for precise 3D imaging. Primarily used by officials in tennis, soccer, and other sports for review calls, it's increasingly used in baseball to help track the throwing motion of pitchers.
This is important for two reasons. First, a successful pitcher is able to replicate the same throwing motion over and over again. Second, close analysis of a pitcher's biomechanics can help find likely causes of injuries.
But Hawk-Eye cameras are only accessible to the home team at MLB and triple-A level ballparks. Any minor-league team at a lower level or a scout at a college or high school field doesn't have access to those analytics because the expensive infrastructure is almost always unavailable.
Winnipeg Free Press | Newsletter
Winnipeg Jets Game Days
On Winnipeg Jets game days, hockey writers Mike McIntyre and Ken Wiebe send news, notes and quotes from the morning skate, as well as injury updates and lineup decisions. Arrives a few hours prior to puck drop. Sign up for The Warm-Up
That's where PitcherNet, a joint project between the Orioles and Bright and Waterloo professor John Zelek, comes in.
'The whole premise of the project was to replicate those capabilities from just using your smartphone by a scout sitting in the stands,' said Zelek a systems design engineering professor at Waterloo and co-director of the university's Vision Image Processing lab. 'So rather than the scout reporting qualitatively, now you have quantitative data from the scout at a college-level game, or wherever.'
Although similar products exist, the proprietary rights to PitcherNet will be owned by the Orioles and, once it's operational, they can explore using it in more ways.
'So much of this game is using your body to generate leverage, whether you're a pitcher to throw the ball or whether you're a hitter to swing the bat,' Hejdal said in a video call from his office at Oriole Park in Camden Yards. 'We had to start somewhere, and the pitching seemed the obvious starting point.'
This report by The Canadian Press was first published May 6, 2025.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
CTV News
19 hours ago
- CTV News
Vending machines at the University of Waterloo violated privacy act
Ontario's privacy watchdog has ruled vending machines on campus at the University of Waterloo violated the Freedom of Information and Protection of Privacy Act. Students lodged a formal complaint with the Office of the Information and Privacy Commissioner of Ontario (IPC) in Feb. 2024 after the 'smart' vending machines were installed by a third-party provider. They claimed the machines were using facial recognition technology to collect images without consent or notice. The students said they were alarmed when one of the machines, located in the Modern Languages building, malfunctioned and displayed an error message reading, ' – Application Error.' 'We wouldn't have known if it weren't for the application error. There's no warning here,' said River Stanley, a fourth-year student at UW, who investigated the machines for an article in the university publication, mathNEWS. Stanley told CTV News students then started putting sticky tack, chewing gum, or Post It Notes over the machines' sensors. vending machine facial recognition Fourth-year University of Waterloo student River Stanley explains where students have been trying to cover a hole on a vending machine that they believe houses a camera. (Colton Wiens/CTV Kitchener) The investigation An IPC investigator learned the university had signed an agreement with the company that owns the machines, Adaria, in Oct. 2023. Adaria was to provide 29 vending machines and be responsible for maintaining, monitoring and stocking them. They were installed in Dec. 2023 and removed from campus in Feb. 2024 when the school learned of the privacy concerns. The university said Adaria either purchased or leased the machines from candy maker MARS and MARS contracted Invenda to build and supply the machines. The school told the IPC it had no knowledge that facial detection technology was being used to collect demographic data. According to an IPC report, when a sale was made, the machine would record a timestamp, the item purchased and demographic data, including facial detection. The technology would then estimate the buyer's gender and age range. 'There was no dispute that the IVMs [Intelligent Vending Machines] captured video images of individuals' faces on the university's campus,' the report read. 'However, the university argued that the resolution of the optical sensor in the IVMs was too low for the device to be considered a camera or create identifiable images of individuals.' The IPC said an investigator deemed the images were of 'photographic quality' but noted those images were held for milliseconds before they were converted into abstract grayscale images and then into numeric descriptors describing the demographic data. 'Our investigation into this matter has found no evidence to suggest that personal information, beyond the initial temporary capture of facial images, was retained and further used by these vendors,' the report said. vending machine facial recognition A vending machine at University of Waterloo displays a facial recognition app error. (Reddit) Did the university know? The IPC report noted the agreement between the University of Waterloo and the vending machine company contained 'all the appropriate standard clauses necessary to protect personal information.' It also determined the university was not aware the machines had facial detection technology that was collecting personal information and had not asked for vending machines with that capability. However, Adaria's proposal did mention a collaboration with MARS to test new product innovations, including MARS Intelligent Vending Machines. The investigators stated that, although the university had reasonable contractual measures in place, it failed to carry out the necessary due diligence that would have uncovered the potential privacy concerns. Report recommendations The IPC report concluded with two recommendations: the university should review its privacy policies to ensure any future collection of personal information complies with the Freedom of Information and Protection of Privacy Act, and the university should ensure it carries out all necessary due diligence to identify, assess and mitigate any potential risks to personal information when entering into new agreements with third-party providers.
Winnipeg Free Press
2 days ago
- Winnipeg Free Press
Tigers lefty Skubal laments 2 walks in otherwise impressive 4-1 win over Orioles
BALTIMORE (AP) — Seven shutout innings. Only three hits allowed, along with two measly walks. Almost every pitcher in the big leagues would be delighted with those results. Not Detroit Tigers lefty Tarik Skubal, who wasn't fully satisfied with his performance in a 4-1 victory over the Baltimore Orioles on Thursday night. 'I was kind of fighting myself early, and as the game wore on, I felt like I got better,' Skubal said. 'I'm probably more proud of grinding it out than when you're on early.' On a hot, humid night at Camden Yards, the 2024 AL Cy Young Award winner looked rather ordinary in the first inning. Baltimore's leadoff hitter, Jordan Westburg, whacked a drive to right that was caught by leaping Kerry Carpenter at the wall. Two batters later, Gunnar Henderson singled. Skubal then walked Coby Mayo in the second inning and Ramón Laureano in the seventh. The two free passes were the most he'd allowed since April 2. 'During my next bullpen,' Skubal said with a wry smile, 'I'll go work on my command.' When you're almost always sharp and that overpowering, it's difficult tough to cope when things don't go perfectly. Skubal (7-2) threw 98 pitches, 74 in the strike zone, but those 24 other pitches caught his attention. 'I just think my misses were kind of bigger than they usually are,' he said. Tigers manager AJ Hinch was impressed with the results, but acknowledged that Skubal wasn't quite as impressive as usual. Thursdays Keep up to date on sports with Mike McIntyre's weekly newsletter. 'He didn't quite have the command he normally has,' Hinch said. 'That doesn't mean he had bad command tonight. It just means it was sort of not perfection. If that's your off day and it's seven scoreless, that's pretty good pitching.' With his sixth scoreless start of the year, Skubal lowered his ERA to 1.99. Yet, he was off his game just enough to draw a good-natured quip from his manager. 'He did walk a couple guys,' Finch noted, 'so we're going to have him work on some things.' ___ AP MLB:
CTV News
2 days ago
- CTV News
University of Waterloo researchers look into what keeps hockey fans hooked
Researchers at the University of Waterloo want to know what happens to the brain when fans watch hockey. CTV's Heather Senoran reports. Loading the player instance is taking more time than usual Loading the player instance is taking more time than usual Researchers from the University of Waterloo are monitoring brain activity to find out what keeps hockey fans hooked on watching their favourite sport. While watching a game, the brains of die-hard hockey fans reacted differently compared to casual viewers, according to their study. The study involved 20 University of Waterloo students who watched the first period of a 2018 European Hockey League game while hooked up to a non-invasive device known as a functional near-infrared spectroscopy (fNIRS). 'We didn't want to show a Leaf's playoff game or anything like that because presumably even the more casual fans might already have preconceived notions,' said Luke Potwarka, a professor and co-author of the study. Researchers analyzed the brain activity of participants wearing the device. 'So based off of these different channels in the headband, we can see as people are say sitting watching a sport game, how blood flow is changing in different parts of the brain,' said Adrian Safati, PhD candidate the Department of Psychology. More blood flow usually means more excitement, even if there isn't much going on in the game. 'There was actually no scoring. And we carefully intentionally chose that game because presumably that key moment would have been the same to both,' said Potwarka. Researchers found that it wasn't scoring opportunities that caught their attention but the brain activity during offensive faceoffs. Researchers said those who were hockey fanatics were more engaged during those times compared to non-committed viewers of the game. 'They see something in the works in motion and they have this greater understanding of how that might be unfolding,' said Safati. This was the first study from the Spectator Experience and Technology Laboratory at the department of Recreation and Leisure Studies. The team said it won't be their last. 'We want to see how does it look in basketball, what are the key moments? What are the moments of active activity and engagement that us viewers might have and those different types of sports?' said Potwarka.
