/Patent Tracker
GoogleGranted patentAI & softwareInterest 6/10

Google patents AI system that filters out people from generated images

This patent describes a method where an AI system creates images based on specific categories while strictly avoiding depictions of people. The system filters out unwanted prompts and generated pictures containing excluded objects before storing them. When a user requests content in that category, Google sends a digital component built from these pre-approved, person-free images.

Why it mattersIt shows how tech companies are building automated safety layers into AI image generation to ensure specific content rules, like excluding people, are met before images are shown to users.

In the filing’s own words

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for automated image generation. Methods can include generating, by an artificial intelligence (“AI”) system, image prompts based on (i) a category and (ii) a set of constraints. At least one of the image prompts is removed based on a set of excluded terms that are not allowed to be included in qualified image prompts. A set of images is generated using the set of qualified image prompts. One or more images that include excluded objects are removed to obtain a set of qualified generated images. The set of qualified generated images are stored with data specifying the category. At least one qualified generated image is combined with other content, resulting in a new digital component. The new digital component is transmitted to a client device in response to a request for content in the category.

What exactly they claim (claim 1)

1. A method, comprising: generating, by an artificial intelligence (“AI”) system, image prompts based on (i) a category and (ii) a set of constraints that instruct the AI system to generate the image prompts in a manner that causes depictions of people to be omitted from generated images that are generated using the image prompts; removing, from the image prompts generated by the AI system, at least one of the image prompts based on a set of excluded terms that are not allowed to be included in qualified image prompts, wherein removal of the at least one of the image prompts results in a set of qualified image prompts; generating, by the AI system, a set of generated images using the set of qualified image prompts; removing, from the set of generated images, one or more images that include excluded objects that are not allowed to be included in qualified images, wherein removal of the one or more images results in a set of qualified generated images; storing, in a data structure, the set of qualified generated images with data specifying the category, wherein storing the set of qualified images with data specifying the category further comprises storing each qualified image in the set of qualified images with (i) data specifying the category and (ii) dimensions of the qualified image; combining at least one qualified generated image from among the set of qualified generated images with other content, wherein the combining results in a new digital component; receiving a request for content in the category; transmitting, in response to the request, the new digital component to a client device.

Facts

Company
Google (Google LLC)
Official title
Generative artificial intelligence
Patent number
US12731210B2
Application number
18629476
Filed
Apr 8, 2024
Granted
Sep 8, 2026
Inventors
3
Claims
18

Classified under

G06T 11/00 Physics & computingG06T 3/403 Physics & computingG06V 10/74 Physics & computingG06F 40/284 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.