Content Identification via Cross-Matching User Fingerprints

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current content identification methods using third-party fingerprint identification services are costly and prone to errors due to resource constraints and deficiencies in matching quality, particularly when identifying video or audio content on communication devices.

Innovation Solution

A system that employs cross-matching of content fingerprints between communication devices and reference data to group users and identify content, utilizing both internal and external identification services to reduce costs and improve matching accuracy by propagating content identification within user groups.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If third-party fingerprint identification services are used for content identification, then content identification can be performed, but identification costs increase and matching accuracy decreases

Engineering Contradiction:

Improvematching accuracyVSAvoididentification costs

Core Design Contradiction:

Measurement precisionVSLoss of energy

Solution Approach 1:

The patent combines internal fingerprint matching capabilities with selective third-party service usage. The system merges the strengths of both approaches by performing initial matching internally and only invoking third-party services when internal matching is insufficient, thereby reducing costs while maintaining accuracy.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The content identification process is segmented into multiple stages: initial fingerprint extraction, internal matching against reference database, confidence level assessment, and conditional third-party verification. This segmentation allows the system to handle most cases internally while selectively using external services only when needed.

Inventive Principle:

Principle #1Segmentation

2Reliability

If third-party fingerprint identification services are used, then content identification can be performed, but the system becomes more prone to errors due to resource constraints

Engineering Contradiction:

Improveidentification reliabilityVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary fingerprint extraction and internal matching before invoking third-party services. This preliminary action filters out most cases that can be resolved internally, reducing the burden on third-party services and minimizing errors related to resource constraints.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements a feedback mechanism where third-party service results are used to validate and refine internal matching algorithms. The confidence level assessment based on matching quality provides feedback that helps the system learn and improve its internal matching capabilities over time.

Inventive Principle:

Principle #23Feedback

3Productivity

If cross-matching between users is implemented, then identification costs reduce and accuracy enhances, but user data processing complexity increases

Engineering Contradiction:

Improveidentification efficiencyVSAvoiddata processing complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The fingerprint reference database serves multiple functions: it stores individual user fingerprints for direct matching, maintains group fingerprints for cross-user matching, and provides reference data for confidence level assessment. This multi-functionality enables cost-effective cross-matching without requiring separate systems for each purpose.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system uses fingerprint templates and confidence level thresholds as intermediaries to manage the complexity of cross-user matching. These intermediaries abstract the complex data processing requirements, allowing the system to perform efficient matching while hiding the underlying complexity from users and external services.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11727045B2Using cross-matching between users and matching against reference data to facilitate content identification

Publication Date: 2023.08.15 GOOGLE LLC

AI summary

Techniques for identifying video and/or audio content are presented. An identification management component (IMC) groups users determined to be at least potentially be watching the same content based on first content fingerprints associated with respective content viewed by the users on communication devices, and a defined matching criterion. IMC facilitates identifying the content associated with the group of users based on first content fingerprints associated with the group of users and first reference content fingerprints. IMC receives a second content fingerprint associated with a communication device of a user in the group, sends the second content fingerprint to a third-party identification component for comparison with second reference content fingerprints, and receives content identification information in response. IMC identifies the content for the group of users based on the content identification information and/or the content identification made based on the first content fingerprints associated with the group.