Image & Video Processing SDK for Tegra X2 GPU

Updated: September 17, 2026

NVIDIA Tegra X2 is the latest mobile processor with high-performance Maxwell GPU architecture. Tegra X2 offers very high CPU and GPU performance to mobile computing. PC class imaging applications that require high performance, low energy consumption and large amounts of memory can now be developed for mobile devices with Tegra X2. We have ported our high performance Image & Video Processing SDK to Tegra and now we can offer super fast imaging solutions for Nvidia Tegra X2 GPU for real-time imaging and video applications.

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 files, offline), both built on our GPU RAW Processor pipeline. The benchmark module is included in the free builds of both applications.

Detailed benchmark data. Measured results are published on our benchmark pages: Jetson / Xavier benchmarks, all benchmarks.

Choosing a Jetson generation? This page covers the earlier NVIDIA Jetson TX2 (Tegra X2, Pascal). For current modules use Jetson Orin ISP SDK or Jetson Xavier SDK. Other earlier: Jetson TX1 (Tegra X1), Jetson TK1 (Tegra K1).

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.

Tegra X2 benchmarks for image and video processing

What image processing performance does the Jetson TX2 deliver?

We have done measurements of kernel times for the most important components of GPU image and video processing SDK from Fastvideo. We utilized images with 2K and 4K resolutions and got the following averaged results. This is just a small set of features from what we have in our SDK.

Tegra X2 performance for 2K images (1920×1080)

HQLI Debayer (8-bit, RGGB) – 0.53 ms
HQLI Debayer (16-bit, RGGB) – 0.66 ms
DFPD Debayer (8-bit, RGGB) – 2.3 ms
DFPD Debayer (16-bit, RGGB) – 1.9 ms
MG Debayer (16-bit, RGGB) – 6.1 ms
JPEG Encoder (8-bit, quality 90%) – 1.2 ms
JPEG Encoder (24-bit, quality 90%, 4:2:0) – 2.0 ms
JPEG Encoder (24-bit, quality 90%, 4:4:4) – 3.1 ms
JPEG Decoder (8-bit, quality 90%) – 2.5 ms
JPEG Decoder (24-bit, quality 90%, 4:2:0) – 5.3 ms
JPEG Decoder (24-bit, quality 90%, 4:4:4) – 6.1 ms
24-bit image Resizer (algorithm Lanczos3) from 1920×1080 to 960×540 – 4.9 ms
24-bit image Resizer (algorithm Lanczos3) from 1920×1080 to 1919×1079 – 9.9 ms
Denoiser (8-bit, wavelet 9/7, 7 dwt resolutions) – 4.4 ms
Denoiser (24-bit, wavelet 9/7, 7 dwt resolutions) – 11.3 ms
J2K Encoder (24-bit, wavelet 9/7, 7 dwt resolutions, cb 32, cr 12, lossless, single) – 118 ms
J2K Encoder (24-bit, wavelet 9/7, 7 dwt resolutions, cb 32, cr 12, lossy, single) – 57 ms
J2K Encoder (24-bit, wavelet 9/7, 7 dwt resolutions, cb 32, cr 12, lossy, batch) – 54 ms

Tegra X2 performance for 4K images (3840×2160)

HQLI Debayer (8-bit, RGGB) – 2.15 ms
HQLI Debayer (16-bit, RGGB) – 2.6 ms
DFPD Debayer (8-bit, RGGB) – 9.6 ms
DFPD Debayer (16-bit, RGGB) – 7.5 ms
MG Debayer (16-bit, RGGB) – 25 ms
JPEG Encoder (8-bit, quality 90%) – 5.0 ms
JPEG Encoder (24-bit, quality 90%, 4:2:0) – 7.9 ms
JPEG Encoder (24-bit, quality 90%, 4:4:4) – 12.5 ms
JPEG Decoder (8-bit, quality 90%) – 9.1 ms
JPEG Decoder (24-bit, quality 90%, 4:2:0) – 19.5 ms
JPEG Decoder (24-bit, quality 90%, 4:4:4) – 21.4 ms
24-bit image Resizer (algorithm Lanczos3) from 3840×2160 to 1920×1080 – 17.7 ms
24-bit image Resizer (algorithm Lanczos3) from 3840×2160 to 3839×2159 – 35.6 ms
Denoiser (8-bit, wavelet 9/7, 7 dwt resolutions) – 16.7 ms
Denoiser (24-bit, wavelet 9/7, 7 dwt resolutions) – 43 ms
J2K Encoder (24-bit, wavelet 9/7, 7 dwt resolutions, cb 32, cr 12, lossless, single) – 442 ms
J2K Encoder (24-bit, wavelet 9/7, 7 dwt resolutions, cb 32, cr 12, lossy, single) – 184 ms
J2K Encoder (24-bit, wavelet 9/7, 7 dwt resolutions, cb 32, cr 12, lossy, batch) – 179 ms

Tegra X2 performance for 12-bit per pixel images 4032×2192

JPEG Encoder (gray, 12-bit, quality 90%) – 8.5 ms
JPEG Encoder (color, 12-bit, quality 90%, 4:2:0) – 13.4 ms
JPEG Encoder (color, 12-bit, quality 90%, 4:4:4) – 22.4 ms

Benchmarks for Fastvideo SDK on Tegra K1, X1, Xavier

Roadmap

Embedded World, Multi-camera embedded system on Tegra X2 - done
Fastvideo SDK benchmarks on Tegra X2 - done
CUDA JPEG2000 Encoder - done
CUDA JPEG2000 Decoder - done

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.