Adobe Patent for Color Conditioned Diffusion Prior Image Processing

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 24th, 2026

Detected March 25th, 2026

Summary

The USPTO has granted Adobe Inc. a patent for a novel image processing system that utilizes color-conditioned diffusion priors. This technology enables the generation of images based on both text and color prompts, enhancing creative control in AI-driven image synthesis.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has granted patent US12586271B2 to Adobe Inc. for a "Color conditioned diffusion prior" image processing system. The patent describes a method where a multi-modal encoder processes text and color prompts to generate embeddings. A diffusion prior model then creates an image embedding, which a latent diffusion model uses to generate an image that incorporates elements from both the text and color prompts. This represents a new technological advancement in AI-powered image generation.

This patent grant signifies a new intellectual property asset for Adobe in the field of AI and image processing. While not imposing direct compliance obligations on other entities, it highlights a specific technological approach that competitors or licensees may need to consider. Companies involved in AI image generation, particularly those utilizing diffusion models, should be aware of this patented technology and its potential implications for their own development and product offerings.

Source document (simplified)

← USPTO Patent Grants

Color conditioned diffusion prior

Grant US12586271B2 Kind: B2 Mar 24, 2026

Assignee

ADOBE INC.

Inventors

Pranav Vineet Aggarwal, Venkata Naveen Kumar Yadav Marri, Midhun Harikumar, Sachin Madhav Kelkar, Hareesh Ravi, Ajinkya Gorakhnath Kale

Abstract

Systems and methods for image processing are described. Embodiments of the present disclosure, via a multi-modal encoder of an image processing apparatus, encodes a text prompt to obtain a text embedding. A color encoder of the image processing apparatus encodes a color prompt to obtain a color embedding. A diffusion prior model of the image processing apparatus generates an image embedding based on the text embedding and the color embedding. A latent diffusion model of the image processing apparatus generates an image based on the image embedding, where the image includes an element from the text prompt and a color from the color prompt.

CPC Classifications

G06T 11/60 G06T 19/00 G06T 11/00 G06T 11/20 G06T 11/206 G06T 7/0012 G06T 2207/10081 G06T 2207/30004 G06T 11/203 G06T 2211/441 G06T 2210/28 G06T 11/001 G06F 9/4443 G06F 3/0481 G06F 30/13 G06F 3/04845 G06F 3/04883 G06F 40/143 G06F 16/54 G06F 16/56 G06F 16/5838 G06F 16/58 G06F 16/55 G06F 40/40 G06V 20/20 G06V 10/7753 G11B 27/10 H04N 21/47217 G16H 30/20 G06N 20/20 G06N 3/045 G06N 3/047 G06N 3/08 G06Q 30/0643