Image-Based Product Search with Category Tree
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search systems are inefficient in finding a specific object when the user lacks precise description, leading to cumbersome navigation through numerous search results, as they fail to utilize additional visual information and often present results in paginated formats, making it time-consuming to locate the desired product.
Innovation Solution
A system that constructs a category tree based on object attributes and uses graphical depictions to match user-input images with corresponding categories, allowing for faster navigation and reducing the search space by presenting all results on a single page, enabling users to quickly find desired products with minimal input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional search systems present all search results in paginated formats, then the system can handle large volumes of search results, but the user experience deteriorates due to time-consuming navigation through multiple pages
Solution Approach 1:
The patent segments the search results into a hierarchical category tree structure, where results are organized by product categories and subcategories rather than being presented as flat paginated lists. This allows users to navigate through categorized groups efficiently, reducing the time spent searching through large volumes of results while maintaining the ability to handle extensive product inventories.
2Measurement precision
If the search system requires precise textual descriptions to find specific objects, then the accuracy of search results improves, but the ease of operation deteriorates because users cannot easily describe visual properties
Solution Approach 1:
The patent replaces the mechanical/text-based description system with an image-based visual search system. Instead of requiring users to type precise textual descriptions of object properties, the system allows users to upload images that are then processed through image recognition and comparison algorithms, automatically extracting visual features and matching them against the category tree, thereby maintaining high accuracy while dramatically improving ease of operation.
3Loss of information
If the search system displays detailed descriptions of all search results, then the completeness of information improves, but the device complexity increases due to the need to process and display extensive text content
Solution Approach 1:
The patent extracts and displays only the most relevant visual and categorical information from search results, rather than presenting complete textual descriptions of all products. The system extracts key visual features from images and essential category attributes, presenting them in a condensed format that maintains information completeness for decision-making while significantly reducing the complexity of information processing and display.
Data Source
AI summary
The present application describes performing a user initiated search query comprising receiving user input comprising description details of at least one desired object, retrieving a plurality of objects from a database sharing one or more of the description details of the user input, retrieving an image of the at least one desired object based on one or more of the plurality of objects, generating a contour of the image and comparing the generated contour with other related contours of other images stored in the database, displaying all of the available contours of all of the images that match the generated contour, receiving a selection of one of the available contours from the user and performing the search query based on the user selected contour.