Image-Based Retrieval System with Contextual Condition Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in retrieving desired or similar products when they lack information beyond images, as existing systems rely on text input or pre-selected image options, limiting the ability to find matching or complementary items.
Innovation Solution
A retrieval support system that includes a commodity information database for image characteristics, a combination information database, and units for similar item and combination retrieval, allowing users to input images and set conditions like time, place, and occasion to find matching or complementary products.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If users input text data or select from pre-prepared images for retrieval, then the retrieval system can process queries, but users cannot retrieve desired objects when they lack information beyond images
Solution Approach 1:
The patent replaces text-based or menu-based retrieval interfaces with image-based retrieval. Users can upload or capture images of objects they want to find, and the system processes these images to retrieve similar or related items. This substitution allows users to retrieve objects even when they have no textual information about them, effectively resolving the contradiction between information loss and ease of operation.
2Adaptability or versatility
If the system provides only basic retrieval functionality, then the system is simple to operate, but it cannot find similar or complementary items effectively
Solution Approach 1:
The patent implements a multi-functional retrieval system that can perform multiple types of retrieval operations: finding similar items based on image characteristics, finding complementary items that go well together, and providing shopping support functions. The system uses image characteristic information and combination information databases to enable these diverse functions within a unified framework, achieving versatility without excessive complexity.
Solution Approach 2:
The patent employs a nested database structure where the commodity information database contains basic item data, and the combination information database contains nested information about complementary items and relationships. This nested organization allows the system to provide basic retrieval functionality while also offering advanced features like finding complementary items, effectively managing complexity while enhancing adaptability.
3Measurement precision
If users search without condition filters, then the search is quick and simple, but the retrieval accuracy for specific contexts is reduced
Solution Approach 1:
The patent implements condition setting functionality that allows users to pre-specify retrieval conditions such as time, place, and occasion before performing the actual retrieval. The system stores these conditions and applies them automatically during retrieval operations. This preliminary action enables accurate context-specific retrieval without requiring users to repeatedly set conditions for each search, effectively reducing the time loss while maintaining high retrieval accuracy.
Data Source
AI summary
A retrieval support system includes a database which stores commodity image data; a retrieval unit which acquires commodity image data having image characteristic information which is the same as or similar to image characteristic information indicating a characteristic in an image of input image data, from the database, with respect to the input image data; and a condition setting unit which sets a retrieval condition including at least one of time information relating to time, place information relating to places and occasion information relating to occasions. The retrieval unit performs retrieval of the commodity image data based on the retrieval condition set by the condition setting unit. The commodity image data acquired by the retrieval unit and information relating to a commodity different from a commodity corresponding to the commodity image data are output together.