Circle the Player: Q&A with Linda Skovmand and Signe Bering

The technology works by identifying specific players throughout complex game scenarios, applying a visual indicator that follows their movements. By combining advanced player tracking with a distinctive visual language, Circle the Player allows teams to identify and analyze individual performances within the broader context of the game, even more easily than before.

Circle the Player has been rolled out in spring 2025. To learn more about this new feature, we sat down with Linda Skovmand (LS), Software engineer, and Signe Bering (SB), Product designer, both at Veo, and asked about the development process.

What inspired the development of Circle the Player?

LS: Honestly, this has been a dream of mine for years. Way back, in 2021, I started experimenting with drawing big red circles under the players in our interactive video player – just playing around with the idea of spotlighting someone visually. It felt like something we should be able to do in the follow-cam too, and it quickly turned into a side project I would work on whenever I had time, and when I wasn’t too frustrated about all the puzzle pieces not fitting together.

Back then, we didn’t really know what we’d use it for. But then Player Spotlight launched, and this little experiment suddenly felt relevant again. I’d actually shelved the whole thing at that point, but with help from Veo’s AI team, we decided to give it another go. And this time, everything finally started coming together.

SB: From the product side, Player Spotlight was all about helping coaches and players find individual moments faster, without scrubbing through entire matches. But as powerful as that is, we knew something was missing: when watching a moment, it wasn’t always clear who to focus on, especially when several players were close together.

That’s really where Circle the Player came in. It was the natural next step to make the Spotlight experience feel more complete and remove friction. And since Linda had already been exploring this as a side project, we didn’t have to start from scratch – we already had a sense of what was possible.

In our product design team, we’re always looking for ways to reduce the time it takes to get value from video, and this felt like one of those “small feature, big impact” moments. It also taps into a pattern people already know, like in televised sports or video games, where it’s natural to have a visual cue that helps guide your attention. So it made sense, solved a real user need, and built on a feature we’d already seen provide a lot of value.

How did you approach the technical challenge of creating a telestration system that feels as intuitive as the manual highlighting viewers are accustomed to seeing?

LS: Once we decided to restart the project, I broke it down into small milestones. The idea was to move step by step until we got the circle to “stick” to a player in a way that felt reliable and accurate.

The first version was super basic – just highlighting the center of the pitch in a paused video. From there, I made it stick to the center while the video was playing, then upgraded that to stick to a player’s position based on radar data, and finally made it highlight the correct player in a Player Moment, using the actual designs we’d made for the final product.

Sounds straightforward, but there were so many challenges along the way. For one, the math – converting a player’s position on the field into a position on the screen – is complicated. Many matrix functions are involved, and even tiny typos could make the circle end up in a completely wrong place. Luckily, we had working calculations in another language, so I just compared results one step at a time until they matched.

Another tricky part was frame accuracy. If you’re even one or two frames off, the follow-cam might have already changed direction or zoomed in, and suddenly the circle isn’t where it should be. I approached this problem by initially using follow-cam videos with barcodes embedded in them so I could read the exact frame number and be sure of what I was seeing. We obviously didn’t want to show barcodes to users, but they were super helpful during testing while we looked for a cleaner solution.

And then finally, when we switched from test data to real production footage, things broke again. Circles that had been sticking perfectly in tests suddenly started drifting or landing in the wrong place. It turned out there were subtle differences between the raw data and the processed video, like stabilisation or visual edits we do after the camera direction is calculated. That’s where the Veo AI and processing teams came in, and together we figured out how to close that gap. In the end, it was a huge cross-team effort, which was really cool to be a part of.

How did your team balance making the player highlighting both functionally effective and visually distinctive as a Veo feature?

SB: From day one, functionality was non-negotiable. If the circle didn’t show up clearly on the pitch, then nothing else really mattered. We tested a lot, on dozens of recordings with different lighting conditions and pitch colours. That’s why we ended up offering two colour options: green and white. It gives the user a fallback if one doesn’t show up well.

But we also saw this as a branding opportunity. A lot of people share their moments as vertical videos on social media, where our watermark often gets cropped out, so we asked ourselves: could the highlight itself carry some Veo identity? I started referencing shapes from our camera design, using animations and brand colour tones to make it “distinctly Veo”, while including our branding team along the way.

At the same time, we had to accept that design is subjective. What looks great to one person might not land with another. So I kept the visual styling clean and simple enough to work across use cases, and flexible enough to evolve later, like adding more context around the player or layering in future features.

Player tracking presents unique challenges in team sports with constant movement and player overlap. Can you explain how the technology identifies and maintains focus on the correct player throughout various game scenarios?

LS: That’s all thanks to Veo’s AI tracking. Our system tracks what’s happening in the game – ball, players, referees – at all times. When we know both a player’s position and jersey number, we can track that specific player over time and break their movements into what we call “tracklets”. From there, we filter out the moments that are most relevant, based on how close they are to the ball, how long they had it, and so on.

The new thing with ‘Circle the Player’ is that since we know where that player was on the field at a specific time, we can convert that to a screen position – and that’s what lets us draw the circle in the right place in the video. The tech behind it is pretty amazing. We just had to figure out how to turn that data into something visual and helpful for the user.

How do you envision "Circle the Player" contributing to the overall Player Spotlight experience?

SB: For me, it literally puts the spotlight in Player Spotlight. Whether you’re a coach, a parent, or the player yourself, you instantly know who to watch, and that clarity matters. Especially when you’re going through a long playlist of moments, it removes that little bit of mental effort it takes to figure out what’s going on. It might seem like a small detail, but it really helps you stay focused and get more value out of each moment.

It also makes the experience more enjoyable – it adds delight. For players, it’s a tool for self-reflection, to understand movement or positioning better. But it’s also just fun to see yourself highlighted in that way. And if you’re sharing moments with a recruiter or putting together a highlight reel, it helps make sure you’re unmistakably the center of attention.

Circle the Player is available with Player Spotlight add-on. Read more here.

‍