Image Query Mapping Using Entity Scoring for Search Relevance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing search engines struggle to accurately generate relevant search queries based on image inputs, particularly when fine-grained and coarse-grained features are present, leading to suboptimal user experience and search results.

Innovation Solution

A system that annotates query images with labels, identifies associated entities, and generates relevance scores for candidate search queries using a combination of image recognition, knowledge databases, and user context to provide representative search queries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the search engine uses image recognition to identify entities and generate search queries, then the relevance of search results is improved, but the complexity of the system increases

Engineering Contradiction:
Improverelevance of search resultsVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the image processing pipeline into distinct modules: image recognition module for entity identification, query generation module for creating search queries, and scoring module for relevance evaluation. This segmentation allows each module to specialize in specific tasks, improving overall reliability while maintaining manageable system complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary scoring mechanism that evaluates the relevance between identified entities and generated queries before presenting results to the user. This intermediary layer acts as a mediator between the image recognition system and the search results display, filtering and ranking candidates to improve relevance without requiring the entire system to be overly complex.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If the system provides multiple candidate search queries based on image labels, then the user experience is improved, but the time required to process the query increases

Engineering Contradiction:
Improveuser experienceVSAvoidquery processing time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-identifying entities and generating candidate queries during the image processing stage before the user needs to see results. The scoring mechanism is also prepared in advance with relevance criteria, allowing the system to efficiently rank and present multiple candidate queries quickly, thus improving user experience without significant time penalty.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system generates multiple candidate search queries (excessive action) based on different image labels and entities identified in the image. This provides the user with various relevant options, improving ease of operation. The partial action is optimized by only generating queries for the most relevant entities and using efficient scoring algorithms to filter the excessive candidates, balancing user experience with processing time.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS20250384050A1Mapping Images to Search Queries
Publication Date: 2025.12.18 GOOGLE LLC
  • US20250384050A1 patent drawing
  • US20250384050A1 patent drawing
  • US20250384050A1 patent drawing

AI summary

Methods, systems, and apparatus for receiving a query image, receiving one or more entities that are associated with the query image, identifying, for one or more of the entities, one or more candidate search queries that are pre-associated with the one or more entities, generating a respective relevance score for each of the candidate search queries, selecting, as a representative search query for the query image, a particular candidate search query based at least on the generated respective relevance scores and providing the representative search query for output in response to receiving the query image.