Upscaling old footage is one thing, Topaz Labs’ Project Starlight uses diffusion-based AI to turn just about any kind of clip into sharp HD video. This was made possibly by building Project Starlight on an entirely new model architecture, utilizing 6B+ parameters while leveraging the latest NVIDIA hardware,
Diffusion-based AI essentially train themselves by observing a high-quality image, gradually adding more noise, and then learning how it degrades. In this case, it starts with a grainy image and then predicts what the picture looked like before all that noise was added. When Project Starlight launches, users will be able to enhance 10-second clips for free and paid subscribers can upscale 5-minute videos using credits.
- Transform your reality and do everything you love in totally new ways. Welcome to Meta Quest 3S. Now you can get the Batman: Arkham Shadow* and a...
- Explore thousands of unreal experiences with mixed reality, where you can blend digital objects into the room around you or dial up the immersion in...
- Have more fun with friends in Quest. Whether you’re stepping into an immersive game with people from around the world, watching a live concert...
Fully blind restoration: our model automatically denoises, deblurs, upscales, anti-aliases. Simply feed in your footage and watch it transform! No manual tweaks, no parameter tuning.
🎥 Our model can restore footage shot on professional cameras. This example 👇 was filmed using… pic.twitter.com/tmLpZ0Jm0t
— Topaz Labs (@topazlabs) February 6, 2025
Videos degrade due to compression artifacts, blurring, aliasing, noise, atmospheric distortion, missing pixels, etc. Each frame suffers from unique types of corruption, making AI video restoration a highly challenging task. Our technology solves this complexity by analyzing hundreds of frames to accurately restore details, delivering unmatched detail recovery combined with unparalleled temporal consistency,” said Topaz Labs.
[Source]
