Jump to content

18481308. HIGH-RESOLUTION IMAGE GENERATION USING DIFFUSION MODELS (ADOBE INC.)

From WikiPatents

HIGH-RESOLUTION IMAGE GENERATION USING DIFFUSION MODELS

Organization Name

ADOBE INC.

Inventor(s)

Tobias Hinz of Campbell CA US

Siddharth Iyer of San Francisco CA US

HIGH-RESOLUTION IMAGE GENERATION USING DIFFUSION MODELS

This abstract first appeared for US patent application 18481308 titled 'HIGH-RESOLUTION IMAGE GENERATION USING DIFFUSION MODELS

Original Abstract Submitted

Methods, non-transitory computer readable media, apparatuses, and systems for high-resolution image generation using diffusion models include obtaining a prompt and generating, using a first diffusion model, a predicted denoised image at a first resolution based on the prompt. The predicted denoised image is generated at a first intermediate diffusion step of the first diffusion model. The predicted denoised image is upsampled to obtain an upsampled denoised image at a second resolution that is higher than the first resolution. A second diffusion model then generates an output image at the second resolution based on the prompt and the upsampled denoised image.

(Ad) Transform your business with AI in minutes, not months

Custom AI strategy for your specific industry
Step-by-step implementation with clear ROI
5-minute setup - no technical skills needed
Get your AI playbook
Cookies help us deliver our services. By using our services, you agree to our use of cookies.