USPTO Patent Grant US12587623B2 - Image Synthesis

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 24th, 2026

Detected March 25th, 2026

Summary

The USPTO has granted patent US12587623B2 to Koninklijke Philips N.V. for an image synthesis apparatus and method utilizing neural networks. The patent, filed on March 7, 2023, details a system for generating view-shifted images from captured images and 3D spatial data.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has granted patent US12587623B2 to Koninklijke Philips N.V. This patent covers an apparatus and method for image synthesis using a view synthesis neural network. The system receives images and 3D spatial data of a scene, trains a neural network to generate images for different view poses, and generates an audiovisual data stream containing image data, scene data, and trained neural network coefficients.

This patent grant is primarily an intellectual property matter and does not impose direct compliance obligations on regulated entities. However, companies operating in the AI, computing, and digital media sectors should be aware of this granted patent as it pertains to image synthesis technologies, potentially impacting their own research, development, and product offerings.

Source document (simplified)

← USPTO Patent Grants

Image synthesis

Grant US12587623B2 Kind: B2 Mar 24, 2026

Assignee

Koninklijke Philips N.V.

Inventors

Christiaan Varekamp, Bart Kroon

Abstract

A first apparatus comprises a first receiver (301) receiving images of a scene captured and a second receiver (303) receives 3D spatial data for the scene A view synthesis neural network (307) generates view shifted images for the scene for different view poses from the images and the spatial data. A neural network trainer (309) trains the view synthesis neural network (307) based on images of the scene for different view poses. A generator (305) generates an audiovisual data stream comprising: image data for the images, scene data representing the three dimensional spatial data, and coefficient data describing coefficients of the view synthesis neural network (307) after training. A second apparatus receives the audio visual data stream and sets a local neural network (403) based on the coefficient data. The local neural network (403) is then used to generate images of the scene for different view poses.

CPC Classifications

H04N 13/111 H04N 13/172 H04N 19/597 H04N 21/21805 H04N 21/44012 H04N 21/816 H04N 13/156 H04N 13/161 H04N 13/194 G06N 3/02 G06N 3/045 G06T 15/20 G06T 17/00 G06T 19/00