Image Prompt Generation From Regions of Interest
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customers often lack a clear idea of the image they want to create, necessitating multiple iterations with salespersons to find a design direction, and catchphrases are typically created separately, requiring proficiency in constructing instruction sentences for AI text generators.
Innovation Solution
A system and method that allows users to specify a region of interest in an image, generating corresponding text data through an AI text generator, and outputting images using an AI image generator, with saliency analysis to enhance design elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If users directly use AI text generators to generate catchphrases, then text data can be generated automatically, but it requires proficiency in constructing appropriate instruction sentences
Solution Approach 1:
The patent introduces an image processing system as an intermediary between the user and the AI text generator. The system automatically extracts regions of interest from images and generates instruction sentences based on image content, eliminating the need for users to manually construct complex prompts while maintaining high text generation efficiency
2Loss of information
If salespersons generate images multiple times to find design direction, then the customer's image requirements can be understood, but it consumes significant time and resources
Solution Approach 1:
The system implements a feedback mechanism where AI-generated text descriptions of image regions are provided back to users, who can then refine their requirements based on this information. This reduces the need for multiple image generation iterations by providing targeted feedback about what elements should be emphasized or modified in the design
Solution Approach 2:
The system performs preliminary analysis of images by automatically identifying and describing regions of interest before final design completion. This preliminary text generation helps clarify design directions early in the process, reducing subsequent iterations needed to understand customer requirements
Data Source
AI summary
Disclosed is a non-transitory computer-readable recording medium storing a program executable by a computer. The program causes a computer to execute: receiving specification of a region of interest in a first image from a user through a specification means; and generating and outputting text data corresponding to the first image according to the region of interest received in the receiving.