Visual Search Tokens for Dynamic Query Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search techniques, such as keyword searching, are inefficient in helping users discover relevant digital content due to the vast amount of available data, leading to challenges in refining search results effectively.
Innovation Solution
A visual search system that allows users to input initial search terms, dynamically presents matching terms and results, and enables refinement through tokenization, color coding, and interactive selection of additional terms and filters, including action filters, to narrow down search results based on user preferences and historical data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If keyword searching is used to search through vast digital content, then the search can cover a large quantity of data, but the search efficiency and accuracy deteriorate
Solution Approach 1:
The patent replaces traditional keyword-based mechanical text searching with visual image-based search. Users upload images instead of typing keywords, and the system uses image recognition technology to automatically identify and search for visually similar items, dramatically improving search efficiency while maintaining the ability to search through vast digital content repositories
Solution Approach 2:
The patent introduces visual tokens as an intermediary between the user's search intent and the digital content database. These tokens represent visual concepts and attributes that bridge the gap between user preferences and the vast quantity of digital content, enabling efficient matching without requiring manual keyword input
2Measurement precision
If manual keyword input is required for search refinement, then the search can be specific, but the ease of operation deteriorates
Solution Approach 1:
The patent enables the search system to automatically refine results without requiring manual user input. After the initial image-based search, the system automatically presents relevant refinement options (such as categories, attributes, or related items) and can automatically apply refinements based on user selections, eliminating the need for manual keyword input while maintaining high search specificity
Solution Approach 2:
The patent performs preliminary analysis of the uploaded image to automatically generate relevant search categories, attributes, and refinement options before presenting them to the user. This preliminary action prepares the refinement structure in advance, making the subsequent refinement process easier and more intuitive for the user
Data Source
AI summary
Described is a system and method for enabling visual search for information. With each selection of a search term, additional search terms are dynamically selected and presented to the user in conjunction with results matching the currently selected search terms. Likewise, a selected search term may be tokenized and a graphical token presented to the user to represent the selected search term.