/Patent Tracker
GooglePatent applicationAI & softwareInterest 6/10

Google patents method for describing 3D objects using neural networks

Google describes a technique for creating digital 3D models using neural networks that predict how light rays interact with surfaces. The system trains on data points that include ray directions and distances to object surfaces, allowing it to estimate where an object is without needing a full mesh. This approach helps computers understand the shape and position of objects in a scene.

Why it mattersThis method could improve how software renders realistic 3D environments or helps robots understand their surroundings by simplifying how object shapes are calculated.

In the filing’s own words

Systems and methods for generating and using primary-ray based implicit neural representations of three-dimensional objects. In some examples, an implicit neural representation of an object may be generated by training a neural network using training examples that include a reference point, a ray, a perpendicular foot of the ray relative to the reference point, and a distance indicator indicating the shortest distance along the ray between the perpendicular foot and a surface of the object, and, optionally, an intersection indicator indicating whether the ray will intersect with the object. Once trained, the neural network may then be used to generate predictions of whether a ray will intersect with the object, and/or predictions of the shortest distance along a ray between its perpendicular foot and a surface of an object, based on a given reference point, ray, and perpendicular foot of the ray relative to the reference point.

What exactly they claim (claim 1)

1. A computer-implemented method of generating an implicit neural representation of an object, comprising: for each given training example of a plurality of training examples, the given training example including a reference point, a ray, a perpendicular foot of the ray relative to the reference point, an intersection indicator indicating whether the ray will intersect with the object, and a distance indicator indicating a shortest distance along the ray between the perpendicular foot and a surface of the object: generating, using a neural network, an intersection prediction based on the reference point, the ray, and the perpendicular foot of the ray relative to the reference point, wherein the intersection prediction represents a prediction of whether the ray will intersect with the object; generating, using the neural network, a distance prediction based on the reference point, the ray, and the perpendicular foot of the ray relative to the reference point, wherein the distance prediction represents a prediction of the shortest distance along the ray between the perpendicular foot and a surface of the object; comparing, using one or more processors of a processing system, the intersection indicator to the intersection prediction to generate an intersection loss value for the given training example; and generating, using the one or more processors, a distance loss value for the given training example based at least in part on a comparison of the distance indicator to the distance prediction; and modifying, using the one or more processors, one or more parameters of the neural network based at least in part on the intersection loss values and the distance loss values generated for the plurality of training examples.

Facts

Company
Google (Google LLC)
Official title
SYSTEMS AND METHODS FOR GENERATING AND USING PRIMARY RAY-BASED IMPLICIT NEURAL REPRESENTATIONS OF THREE-DIMENSIONAL OBJECTS
Publication number
US20260301304A1
Application number
18881348
Filed
Jul 20, 2022
Published
Oct 1, 2026
Inventors
4
Claims
25

Classified under

G06T 15/06 Physics & computingG06V 10/82 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.