Information processing device, prompt generation method, and image generation method

The information processing device addresses the challenges of using image generation AI by automatically generating prompts for image placement and size based on user input characters and lines, enhancing user convenience and image generation accuracy.

JP2026090881AActive Publication Date: 2026-06-03LENOVO (SINGAPORE) PTE LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
LENOVO (SINGAPORE) PTE LTD
Filing Date
2024-11-22
Publication Date
2026-06-03

AI Technical Summary

Technical Problem

Individuals who are not skilled at drawing or find it difficult to express their ideas through images struggle with conventional image generation AI, leading to anxiety about generating desired images, and the need to explain text placement accurately, which is cumbersome.

Method used

An information processing device that includes an input receiving unit, line detection, character detection, and word cluster determination units to generate prompts for an image generation model based on detected characters and their positions, allowing users to input characters and lines to define image placement and size.

Benefits of technology

Enhances the convenience of using image generation AI by automatically generating prompts that accurately position and size images based on user input, reducing the need for detailed textual explanations and improving image generation accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026090881000001_ABST
    Figure 2026090881000001_ABST
Patent Text Reader

Abstract

To improve the convenience of using image generation AI. [Solution] The information processing device includes: a line detection unit that detects a first input line that spans a first region indicating the range of an image to be input to an image generation model and a second region indicating a predetermined range outside the first region; a character detection unit that distinguishes and detects characters input within the first region and characters input within the second region; a word cluster determination unit that determines a group of characters consisting of one or more words among the detected characters as a word cluster; an extraction unit that extracts a connecting line, one end of which is connected to a word cluster in the second region, and a word cluster connected to the connecting line; a connection destination determination unit that determines the destination of the other end of the connecting line; and a prompt generation unit that generates a prompt as an input sentence to the image generation model based on the word cluster, the connecting line, and the position and size corresponding to the destination of the other end of the connecting line.
Need to check novelty before this filing date? Find Prior Art