MUMBAI, India, April 10 -- Intellectual Property India has published a patent application (202647040615 A) filed by Google Llc, Mountain View, U.S.A., on March 31, for 'sketch-to-image pipeline with automatic prompting.'
Inventor(s) include Osborn, William Roger; Koser, Mikkel Crone; Raento, Mika Petteri; and Lynch, Maura Elizabeth.
The application for the patent was published on April 10, under issue no. 15/2026.
According to the abstract released by the Intellectual Property India: "The present document describes techniques associated with a sketch-to-image pipeline with automatic prompting. These techniques include a method to generate images in a two-stage process using an image-to-text model and a text-to-image model. The input image, which can be a drawn sketch, is processed in order to automatically generate a description of the content. The automatically generated description is combined with a style template to form a prompt. The prompt is used as input to the text-to-image model. The input image is also used as input to the text-to-image model to increase the likeness to the input image. The method simplifies image generation by removing the need for the user of the pipeline to type a prompt (while giving them the option to do so if they wish) and while giving them control over how closely to the input sketch they wish to adhere."
The patent application was internationally filed on May 05, 2025, under International application No.PCT/US2025/027747.
Disclaimer: Curated by HT Syndication.
