/Patent Tracker
GooglePatent applicationAI & softwareInterest 6/10

Google patents simpler AI method to compress images faster

Google describes a way to shrink digital files, like photos, using a simplified version of an AI model called a transformer. Instead of using complex, multi-layered calculations, this method uses a single-scale model with fixed rules to predict how to pack the data efficiently. This approach aims to make the compression process quicker by removing unnecessary steps from the standard design.

Why it mattersFaster file compression could lead to quicker photo uploads and smoother streaming, though this is still just a technical proposal.

In the filing’s own words

Provided are systems and methods that train masked transformer models to perform neural data compression (e.g., image compression). According to one example aspect, instead of relying upon a complex multi-scale model that uses a hyperprior, example implementations of the present disclosure can use a single-scale transformer model that does not use a hyperprior, such as, for example, a standard transformer or similar variants. This greatly simplifies the model architecture and enables improvements in inference speed. According to another example aspect, instead of an uncertainty-adaptive approach recently used for image generation, example implementations of the present disclosure can use predefined, deterministic schedules. The use of deterministic schedules enables the use of masked attention during training (e.g., in addition to masked inputs).

What exactly they claim (claim 1)

1. A computer-implemented method for performing neural data compression with improved efficiency, the method comprising: obtaining, by a computing system comprising one or more computing devices, a plurality of feature tokens associated with a dataset; training, by the computing system, a masked transformer model on the plurality of feature tokens to learn a distribution associated with the plurality of feature tokens, wherein the masked transformer model comprises a single-scale transformer model that does not use a hyperprior; using, by the computing system, the trained masked transformer model to predict one or more distribution parameter values for each of the plurality of feature tokens; and entropy coding, by the computing system, the plurality of feature tokens based on the one or more distribution parameter values predicted by the trained masked transformer model.

Facts

Company
Google (Google LLC)
Official title
Neural Data Compression with Masked Transformer Models
Publication number
US20260279036A1
Application number
19470695
Filed
Mar 28, 2024
Published
Sep 17, 2026
Inventors
3
Claims
22

Classified under

G06V 10/82 Physics & computingG06N 3/0495 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google 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.