NvidiaPatent applicationAI & softwareInterest 6/10

Nvidia patents image resizing that uses math to sharpen details

Patent drawing
Patent drawing from the official filing.

Nvidia describes a method for making digital images larger without losing quality. It uses a neural network to change the image into a mathematical format called the frequency domain, where it can enlarge the picture more effectively before turning it back into normal pixels.

Why it mattersThis technique could help graphics cards display clearer images on high-resolution screens or improve video quality in games and professional software.

In the filing’s own words

Apparatuses, systems, and techniques to scale textured images using a Fourier transform in conjunction with one or more neural networks. In at least one embodiment, a neural network generates an expanded image from an input image by applying a Fourier transform to one or more feature maps generated by said neural network and up-scaling one or more resulting frequency domain feature maps before generating an expanded output image based on up-scaled feature maps.

What exactly they claim (claim 1)

1. A processor comprising: one or more circuits to use one or more neural networks to up-sample pixels of an image after the pixels of the image have been transformed to a frequency domain.

Facts

Company
Nvidia (NVIDIA Corporation)
Official title
FOURIER TRANSFORM-BASED IMAGE SYNTHESIS USING NEURAL NETWORKS
Publication number
US20260359934A1
Application number
19665968
Filed
May 1, 2026
Published
Nov 12, 2026
Inventors
6
Claims
1

Classified under

G06T 3/4084 Physics & computingG06F 18/21347 Physics & computingG06T 3/4046 Physics & computingG06T 3/4053 Physics & computing

More from Nvidia in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.