GPU Image Processing Technologies on NVIDIA CUDA

Updated: July 28, 2026

These are the core image processing technologies — algorithms that run on the NVIDIA GPU — behind Fastvideo products. Each entry links to how it works and to the products that implement it.

Parallel image processing on CUDA

All of these algorithms are parallelized to run entirely on the GPU, chained together in GPU memory without CPU round-trips. See the CUDA image processing library architecture and the Fastvideo SDK.

RAW to RGB conversion (full GPU ISP)

Turning raw sensor data into a finished RGB image on the GPU is the composite technology that chains the steps below — demosaicing, denoising, color and tone mapping — into one pipeline in GPU memory. It is realized in real time in FastVCR camera software and offline in the GPU RAW processor / Fast CinemaDNG Processor.

Codecs (GPU compression)

• GPU JPEG — JPEG image compression on the GPU.
• GPU JPEG2000 — JPEG2000 (J2K) compression on the GPU.
• RAW image compression — lossy and lossless RAW to reduce storage.

Image processing on the GPU

• Deterministic ISP — AI-free, reproducible RAW-to-RGB; no invented pixels.
• Demosaicing — Bayer-to-RGB algorithms (HQLI, DFPD, MG).
• Denoising — real-time NLM and bilateral denoising, no AI.
• Tone mapping / HDR — adaptive local tone mapping.
• Chromatic aberration removal — CA correction in the RAW domain.
• Defringe — removal of color fringing.
• 3D LUT — 2.5D and 3D LUT color transforms.

Looking for a ready product? See all products. For use cases by industry, see applications and solutions.

About the author

Fyodor Serzhenko, PhD, is the founder and CEO of Fastvideo. He earned his PhD at the Moscow Institute of Physics and Technology (MIPT) in 1993. Since 2009 he has led the development of Fastvideo’s GPU-accelerated image codecs and ISP modules, including the Fastvideo SDK. Connect on LinkedIn.

Why you can trust these results

Fastvideo has built GPU-accelerated image processing software since 2009. Our codecs comply with the standards they implement — JPEG (ITU-T T.81 / ISO IEC 10918) and JPEG2000 (ITU-T T.800 / ISO IEC 15444) — and our tools are open source on GitHub. Every performance figure we publish is reproducible: see our benchmark methodology, download the Fastvideo SDK benchmark report (PDF), and measure it on your own GPU.