/Patent Tracker
GooglePatent applicationAI & softwareInterest 6/10

Google patents image creation method that checks its own work

Google describes a system that creates images by breaking them into small data blocks called tokens. A second, separate model scores these tokens to identify which ones look realistic, keeping only the good ones before the system generates the rest of the image.

Why it mattersThis approach could help computer programs create clearer pictures by fixing mistakes in real-time rather than building the image from scratch every time.

In the filing’s own words

Systems and methods for iterative non-autoregressive image synthesis using a first generative model and an independent second token-critic model. In some examples, an image may be synthesized using one or more passes in which the generative model predicts a first plurality of tokens representing a first vector-quantized image, the token-critic model generates a first plurality of scores based on the first plurality of tokens, the processing system selects a first set of one or more tokens of the first plurality of tokens to be preserved based on the first plurality of scores, and the generative model then predicts a second plurality of tokens based on the first set of tokens, the second plurality of tokens including the first set of tokens. In some examples, the generative model may be configured to predict probability distributions, which may be sampled to generate the first and second pluralities of tokens.

What exactly they claim (claim 1)

1. A computer-implemented method of generating an image, comprising: predicting, using a first neural network, a first plurality of probability distributions; generating, using one or more processors of a processing system, a first plurality of tokens based on the first plurality of probability distributions, the first plurality of tokens representing a first vector-quantized image; generating, using a second neural network different than the first neural network, a first plurality of scores based on the first plurality of tokens, each score of the first plurality of scores representing a prediction of whether a token of the first plurality of tokens was generated by a generative model; selecting, using the one or more processors, a first set of one or more tokens of the first plurality of tokens based on the first plurality of scores; predicting, using the first neural network, a second plurality of probability distributions based on the first set of tokens; and generating, using the one or more processors, a second plurality of tokens based on the second plurality of probability distributions and the first set of tokens, the second plurality of tokens including the first set of tokens and representing a second vector-quantized image.

Facts

Company
Google (Google LLC)
Official title
SYSTEMS AND METHODS FOR ITERATIVE NON-AUTOREGRESSIVE IMAGE SYNTHESIS USING INDEPENDENT TOKEN-CRITIC MODEL
Publication number
US20260260391A1
Application number
18993795
Filed
Aug 30, 2022
Published
Sep 3, 2026
Inventors
6
Claims
31

Classified under

G06T 11/00 Physics & computingG06N 3/045 Physics & computingG06N 3/0895 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.