Defringe on CUDA
Updated:
Fastvideo defringe removes color fringing on the GPU — the red-green and blue-yellow fringes from lateral chromatic aberration, and the purple and green fringes around out-of-focus, high-contrast edges. It runs on CUDA inside the raw-to-RGB pipeline, so fringe removal keeps up with real-time camera processing.
Detailed benchmark data. Measured results are published on our benchmark pages: all benchmarks.
Color fringing can result from several physical phenomena
- Lateral (transverse) chromatic aberration (red/green fringes, blue/yellow fringes)
- Axial (longitudinal) chromatic aberration (purple and green fringes)
Our software can reduce chromatic aberrations on CUDA. This is not fully automatic process, so there are some parameters to get better quality of fringe removal. Defringe module could be incorporated in a standard image processing pipeline. That module is a part of Fast CinemaDNG Processor software which is intended for fast DNG processing on CUDA.
Defringe on CUDA
How does GPU defringe remove color fringing?
- Input: 24/48-bit images data from HDD/SSD/RAID in PPM format or from CPU/GPU memory
- Output format: 24/48-bit RGB image to show on monitor in real-time
- High performance defringe and fast OpenGL output
- Command-line application for Windows-10 and Linux Ubuntu
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.