Cinnafilm has announced its collaboration with NVIDIA to produce groundbreaking a advancement in AI image processing. The new technology is designed to process super-resolution HD and UHD upconversions. It will be commercially available later this year inside Cinnafilm’s flagship conversion platform, PixelStrings.
PixelStrings
Cinnafilm has been a leader in GPU-based image processing since 2004. The company also utilizes NVIDIA CUDA architecture to power its award-winning tools, including PixelStrings. PixelStrings is a pay-as-you-go, SaaS (Software as a Service) that operates media conversion workflows in the cloud and on-premises.
In addition, PixelStrings includes the Tachyon app for standards and frame rate conversion.
AI-driven super resolution
The collaboration combines Cinnafilm’s image processing engine with new NVIDIA RTX AI Image Processing. This delivers real-time AI-driven super resolution, based on RTX Video Super Resolution but customized for Cinnafilm. As a result, you can transform low-resolution content into ultra-high-definition video from any source in a single pass.
Improved results
Cinnafilm says that the new AI technology gives quality and detail enhancement gains that are 25-30% greater than its current industry-leading solution. Other upscaling tools rely on generic algorithms or generative AI that can create hallucination errors. However, the new joint solution harnesses the latest NVIDIA Tensor Cores and Cinnafilm’s motion-compensation and noise management. As a result, the process minimizes artifacts, intelligently sharpens and preserves cinematic picture quality.
“Our AI collaboration with NVIDIA represents a major advancement for video upconversion,” said Lance Maurer, CEO and founder of Cinnafilm. “Ensuring our PixelStrings customers have access to safe, effective utility AI tools without introducing the headaches of artifact distortion or adding potential usage risk currently associated with generative AI was paramount. Video engineering is a careful business, and partnering with NVIDIA ensures that our AI strategy builds upon a stable and refined process. We see a bright future with AI in our product sets, and our first foray into AI will be a game-changer for our customers and the wider industry.”
“Advanced speed and scalability are key for accelerating AI image processing,” said Richard Kerris, Vice President of Media and Entertainment at NVIDIA. “By collaborating with Cinnafilm, the new solution pushes the boundaries of what’s possible by leveraging advanced signal processing and AI enhancement, delivering stunning outputs, computational power, and workflow efficiency that will empower creators and enterprises alike.”
What we think
4K streaming and broadcast television is becoming more prevalent around the world. However, there is a lot of still-popular older content that was recorded in standard definition (SD). This can look poor on an HD television, let alone a 4K screen. Resolution upconversion is therefore important to ensure that SD content can be viewed by a modern audience. Hopefully, Cinnafilm’s collaboration with NVIDIA will allow shows from the past to have a new lease of life. It’s also good to bring to you a story that shows the use of AI technology in an unequivocally positive light.
Pricing and availability
Cinnafilm is going to showcase the new technology at the 2025 NAB Show in Las Vegas this month. It will be commercially available later this year inside Cinnafilm’s flagship conversion platform, PixelStrings.
