Jon’s Journey from Sprites to Streams
For the truly non-geek: This piece is aimed at non-technical audiences, too, and I’m told not everyone knows what a “sprite” is. It could be (1) a small magical creature, (2) a tiny, delicate person, (3) a two-dimensional image in computer graphics, (4) a large electrical discharge above thunderstorms, or (5) a lime-flavoured soft drink. Jon usually talks about option 3, but honestly, he brings a bit of all five. Oh, and “real-time” means that computers can do their magic on a video stream without having to delay or slow it (ie, it takes less than an hour to upscale an hour of video – In live scenarios, the magic happens in milliseconds.)
Jon Frydensbjerg didn’t set out to change the future of video. As a teenager in the early 90s, he was chasing something simpler: making pixels move. Tinkering on his Amiga (a personal computer from the 80s), exploring the “demo scene,” Jon fell in love with graphics programming, not for the sake of tech, but for the thrill of conjuring visuals out of raw code.
That obsession with how things look (and how they move) never left him. He’s been coding ever since: first for fun, then for computer vision clients for broadcasters who want their sports feeds to look as crisp as a Pixar render.
Thirty years after tinkering with sprites, Jon’s still working on the same fundamental problem: how to make pixels look better. Only now, he’s doing it at scale, in real-time, and with AI models running in the cloud. As the technical mind behind Pixop, Jon’s early passion for visual quality now powers a tool that helps broadcasters deliver cleaner, sharper, more vivid video, without requiring new gear.
A Brief History of Making Video Look Better
Before video went digital, broadcasters still tried to improve quality by cleaning up noise on VHS tapes and tweaking colour signals, but the real leap came with digital workflows. In the 2000s, engineers started playing with basic enhancement tools: deinterlacing, denoising, and simple upscaling. These worked, but had limits.
That changed around a decade ago. As GPUs got faster and deep learning became accessible, it became feasible to train neural networks on pristine and degraded footage, teaching machines what “good” looks like.
Jon and his team developed specialised AI filters that can enhance both live and archived video. They upscale standard High Definition (HD) into Ultra HD (UHD) by inventing plausible new detail. They convert Standard Dynamic Range (SDR) into High Dynamic Range (HDR) by enriching colour and contrast. And they do it in real time.
What AI-enhancement Actually Does
Jon describes it best: “It’s like a spa treatment for your video.”
Your live video feed, say, HD of mixed quality, gets routed through a small server running Pixop software.
You pick the issues you want to fix: motion blur, blockiness, and washed-out colours.
The right AI models (or “filters”) are applied for those tasks.
In less than a second, the improved video is ready for viewers: sharper, smoother, richer.
Filters are configurable. Jon has even explored combining them into a single adaptive model that adjusts itself to the content, whether it’s football, breaking news, or a poorly-lit ad break.
Won’t Devices Just Catch Up?
Possibly. But Jon’s view is pragmatic. TVs and phones are getting smarter, but they rely on one-size-fits-all tricks that don’t work for everything. Jon and his team at Pixop have trained the models on diverse, real-world footage. They handle edge cases: forest canopies, archive footage, and low-bitrate sports. And they do so upstream, way before the content ever reaches your screen.
This centralised enhancement means:
- The broadcaster controls the quality.
- Viewers get consistency across devices.
- The work is done once, not a million times on a million TVs.
Jon calls that last one a sustainability win.
Doing More With Less
Jon’s small team at Pixop has big ambitions: to help broadcasters meet growing viewer expectations without expensive upgrades. As demands rise (sharper resolution, richer colours, smoother motion), their challenge is to keep quality high, cost low, and complexity invisible.
Whether the future lies in the cloud, on the edge, or inside your TV, one thing seems inevitable: video is only getting better. And Jon’s mission? To make that future more beautiful, without breaking the bank.
Read More:
Jon’s LinkedIn page
