Visual Search Personalization via User Interest Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing visual query systems focus exclusively on identifying images with similar visual characteristics, failing to reflect the user's true search intent and providing limited personalized search results.
Innovation Solution
A computer-implemented method and system that processes visual search queries by obtaining user-specific visual interest data, generating a ranking of search results based on this data, and providing personalized visual result notifications overlaid on the query image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If visual search systems focus exclusively on identifying images with similar visual characteristics, then the system can provide consistent visual matching results, but the search results fail to reflect the user's true search intent and lack personalization
Solution Approach 1:
The system performs preliminary actions by obtaining user-specific visual interest data before processing the visual search query. This pre-acquired data about user preferences is stored and ready for use, allowing the system to quickly personalize results without adding complex real-time analysis during the search process itself.
Solution Approach 2:
The patent introduces user-specific visual interest data as an intermediary element between the visual query and search results. This intermediary data layer mediates the matching process by filtering and ranking results based on user preferences, thereby personalizing outcomes without requiring fundamental changes to the core visual search algorithm.
2Loss of information
If the system provides comprehensive search results for all detected objects in the image, then the user receives complete information, but the user interface becomes cluttered with excessive notifications
Solution Approach 1:
The system applies local quality by providing different levels of result presentation based on user interest. Instead of uniformly displaying all results, the system selectively highlights and prioritizes notifications for objects that match user-specific visual interests, while suppressing or minimizing notifications for less relevant objects. This creates a non-uniform, customized information presentation that maintains completeness while improving interface clarity.
Solution Approach 2:
The patent changes the parameter of result prioritization by introducing user interest weights. Search results are ranked not only by visual similarity but also by their relevance to user-specific interests. This parameter change allows the system to dynamically adjust which results are prominently displayed versus those that are less visible, balancing information completeness with interface cleanliness.
3Measurement precision
If the system returns search results for multiple canonical items, then the user receives comprehensive and accurate results, but the system requires more complex processing to identify and retrieve content for multiple items
Solution Approach 1:
The system performs preliminary actions by pre-identifying and storing user-specific visual interest data that captures the user's preferences for different object types, brands, or categories. This pre-processing allows the system to quickly filter and prioritize results for multiple canonical items during the search execution phase, reducing the computational burden of real-time analysis.
Solution Approach 2:
The patent uses copying by creating and storing user interest profiles that replicate patterns of user preference. Instead of analyzing user behavior in real-time during each search, the system copies these preferences into a reusable data structure that can be efficiently applied across multiple searches, maintaining accuracy while improving processing efficiency.
Data Source
AI summary
A user can submit a visual query that includes one or more images. Various processing techniques such as optical character recognition (OCR) techniques can be used to recognize text (e.g. in the image, surrounding image(s), etc.) and/or various object detection techniques (e.g., machine-learned object detection models, etc.) may be used to detect objects (e.g., products, landmarks, animals, humans, etc.) within or related to the visual query. Content related to the detected text or object(s) can be identified and potentially provided to a user as search results or a proactive content feed. As such, aspects of the present disclosure enable the visual search system to more intelligently process a visual query to provide improved search results and content feeds, including those search results which are more personalized and/or consider contextual signals to account for implicit characteristics of the visual query and/or user's search intent.