Later today (29 April 2025), Intel’s Open Image Denoise will be awarded a Technical Achievement Award from the Academy of Motion Picture Arts and Sciences.
Intel Open Image Denoise is an open source library that provides a high-performance, high-quality AI-based denoising solution for ray traced images.
The Academy, which hosts the annual Oscars, is recognising the library as a contributing innovation in modern film rendering.
“Intel continues to push the envelope on high-quality rendering and visual experiences, delivering benefits to creators around the world, and this achievement is a timely recognition for the passion and persistence of the team,” says Anton Kaplanyan, Intel vice-president: Graphics and GPU Research.
“We are committed to bringing the best visual computing breakthroughs for creators, developers and gamers with our products and software ecosystem, and this top award is a manifest of that coming to life.”
Ray tracing is the foundation of modern rendering. It is a powerful algorithm capable of producing very realistic images, but it has high computational requirements. When using ray tracing on its own to produce noise- and artifact-free images, a large number of rays need to be traced, which is often slow and expensive.
With a sophisticated denoiser like Intel Open Image Denoise, rendering times can be significantly reduced by augmenting the renderer, which enables tracing fewer rays without sacrificing image quality.
Intel Open Image Denoise filters out the unwanted noise inherent to ray tracing using AI neural networks, accelerating real-time previews during the creative process and cutting down final rendering times.
With its simple but flexible C/C++ API, the library can be easily integrated into most existing or new rendering solutions. It also has rich cross-vendor support, with optimizations for most major CPU and GPU architectures from Intel, AMD, Nvidia, Apple and Arm.
Intel Open Image Denoise is part of the Intel Rendering Toolkit and is released under the Apache 2.0 license. Its core technology is provided by the widely adopted U-Net architecture that is highly efficient and preserves details, raising the quality of computer-generated imagery across the industry.
Since the library is open source and free to use, the included training tool kit enables users to train custom denoising models using their own datasets, allowing for more flexibility and higher image quality.
Additionally, creators and studios can also re-train the included denoising neural networks for their own renderers, styles and films.
The library is also integrated into many well-known rendering tools, including Autodesk Arnold, Blender, Chaos V-Ray, Chaos Corona, Foundry Modo, Maxon Cinema 4D and others.
The Academy introduced the Scientific and Technical Awards in 1931 to recognise the achievements in science and technology’s critical role in advancing motion picture production.