CUDA Resizer from Fastvideo

Updated: August 27, 2026

Fastvideo CUDA Resizer is a GPU library that upscales and downscales grayscale and color images with the Lanczos algorithm in a floating-point pipeline. It is fast enough for real-time work — a Full HD 24-bit frame resizes to 960×576 at about 3000 fps on an NVIDIA GeForce GTX 1080 — and ships as a module of the Fastvideo GPU Image & Video Processing SDK.

How these numbers were measured. Fastvideo benchmarks are measured on NVIDIA GPUs, timing pure GPU processing at the resolution, bit depth, format and quality stated on this page. See the full benchmark methodology and test environment and the Fastvideo SDK benchmark report (PDF). Our tools are open source on GitHub and built on NVIDIA CUDA.

Check a single module. Every module in the Fastvideo SDK comes with a command-line sample. Run it on your own images on your own GPU and compare the result with the figures published here.

Check the whole pipeline. A module in the middle of a pipeline cannot be judged on its own: its input is hard to prepare, and its output is an intermediate product you cannot look at. What matters is how it changes the final image and the total time of the schema. Two of our applications run the complete pipeline and report the time of every stage — FastVCR (live cameras or RAW files from disk, so no camera is required) and Fast CinemaDNG (DNG, offline). The benchmark module is included in the free builds of both.

Detailed benchmark data. Measured results are published on our benchmark pages: all benchmarks.

We wrote these algorithms ourselves

Image processing has always been a trade-off between quality and speed. The methods that give a good result take minutes per frame; the fast ones leave halos and artifacts. The reason is the compute budget: on a CPU there was never enough of it, so algorithms were simplified — and much of the industry still runs on simplifications accepted decades ago.

Fastvideo’s codecs and ISP modules are neither ports of that legacy nor wrappers around third-party code. We designed — and in many cases invented — the algorithms behind our demosaicing, denoising, tone mapping and JPEG/JPEG2000 codecs, and wrote them for CUDA from the start, on top of image-processing expertise built since 2009. An algorithm moved to the GPU still carries the compromises it was simplified for; ours were never made under that constraint. On a modern GPU the budget is large enough that, for practical workloads, the choice between quality and speed no longer has to be made.

That is why you get a complete, tuned imaging solution rather than a bag of GPU primitives — and why, when your pipeline needs something specific, we can change the algorithm itself, not just call it.

CUDA Resizer Features

Input images: 8-bit or 16-bit per color component RGB, PGM, PPM, BMP, byte array in CPU/GPU memory
Resize images and video to any size quickly and with high quality
Image resize algorithm for upscale and downscale: Lanczos
Internal calculations with floating point precision
Writes resized images to JPEG, BMP, PPM, PGM, DIB formats
Compatible with NVIDIA GPUs Maxwell, Pascal, Volta, Turing, Ampere, Ada

OS Supported

Windows-10
Linux Ubuntu
Linux4Tegra (L4T)

CUDA Resizer benchmark on NVIDIA GeForce GTX 1080

How fast is GPU image resizing?

High quality image resize for color Full HD (1920×1080, 24-bit) image to final resolution 960×576 can be done on NVIDIA GeForce GTX 1080 GPU at frame rate 3000 fps. We recommend to use our resize solution on CUDA together with other components from our Image Processing SDK to be able to do fast and high quality resize for grayscale or color images.

CUDA Resize for JPEG images

If we need to resize JPEG images on GPU, we need to decode these images first. That could be also done on GPU with the aid of CUDA JPEG Codec. After that we can apply CUDA Resize and some optional transforms on GPU like Crop, Rotation, Sharpening, etc. At the final part of such a pipeline we usually apply JPEG Encoder to get output image in jpg format. Here you can see more info about JPEG Resizer benchmarks on NVIDIA Tesla V100 and review for the latest solutions and benchmarks on GPU and FPGA.

Recently we've got significant performance boost for JPEG Resize with CUDA MPS on Linux. Now such a solution is much faster than CPU and FPGA implementations for JPEG Resize.

Licensing

We license CUDA Resizer and other components of GPU Image & Video Processing SDK to software developers, camera manufacturers, internet providers, software integrators, etc. Our SDK is utilized in wide range of high performance imaging applications. SDK evaluation version, documentation, licensing info and quotation are available upon request. We are also offering custom software design according to agreed specification. If you need to get significant speed up for your image processing application, don't hesitate to contact us.

Roadmap for further improvements of CUDA Resizer

CUDA Resizer integration into Fast CinemaDNG Processor software - done
Minimum memory usage on CUDA applications
More input/output formats for CUDA Resizer
New high quality image resize algorithm on CUDA
Other CUDA Resizer optimizations
High performance GPU Resize for H.264 and H.265 video transcoding
FFmpeg resize filter for video transcoding

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.