Distributed Multimedia Database System for Scalable Identification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current multimedia identification systems face challenges in handling large multimedia databases with high capacity and density, requiring sophisticated algorithms and hardware that exceed the capabilities of existing software solutions, especially in handling distortions and scalability.

Innovation Solution

A distributed and tiered multimedia database system is implemented, with a reference database split into identifiable portions stored across multiple servers for parallel search, using cluster indexes and multimedia signature indexes for efficient querying and caching, and enabling local client tracking and query caching to manage and scale multimedia content identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If a centralized multimedia database system is used, then it is simple to manage, but it cannot handle large database capacity and high density efficiently

Engineering Contradiction:

Improvedatabase capacityVSAvoidsystem complexity

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent divides the large multimedia database into multiple distributed servers, each handling a portion of the database. The reference database is split into identifiable portions stored across multiple search servers, allowing the system to scale horizontally to handle large database capacity while maintaining manageable complexity through modular architecture.

Inventive Principle:

Principle #1Segmentation

2Reliability

If sophisticated algorithms and hardware are used to handle distortions, then identification accuracy improves, but system complexity and cost increase

Engineering Contradiction:

Improveidentification accuracyVSAvoidalgorithm complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent pre-processes multimedia content by extracting features and creating fingerprints before storage. This preliminary action includes generating robust feature representations that are inherently resistant to distortions, reducing the need for complex real-time analysis algorithms while maintaining high identification accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces complex mechanical/hardware-based distortion handling with software-based fingerprinting algorithms. The system uses digital signal processing and pattern recognition techniques to achieve robust identification against distortions, substituting sophisticated hardware requirements with intelligent software solutions.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If the reference database is distributed across multiple servers, then search speed and scalability improve, but system management complexity increases

Engineering Contradiction:

Improvesearch speedVSAvoidsystem management

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent implements a universal query interface that works across the distributed database system regardless of which server holds the relevant data. The system provides multi-functional capabilities including distributed search, result aggregation, and automatic load balancing through a unified interface, simplifying operation while maintaining high search speed.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces a query distribution mechanism that acts as an intermediary between the user and the distributed servers. This intermediary automatically routes queries to appropriate servers, aggregates results, and manages the complexity of distributed system operations, allowing fast parallel search while simplifying system management.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Reliability

If fingerprinting is used instead of watermarking, then content identification is more robust, but the system becomes more challenging to scale

Engineering Contradiction:

Improvedistortion robustnessVSAvoidsystem scalability

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent implements scalable fingerprinting by changing parameters such as the number of features extracted, the dimensionality of feature vectors, and the granularity of database partitioning. These parameter adjustments allow the system to maintain distortion robustness while adapting to different scalability requirements without fundamental architectural changes.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUSRE48791E1Scalable, adaptable, and manageable system for multimedia identification

Publication Date: 2021.10.26 ROKU INC

AI summary

An architecture for a multimedia search system is described. To perform similarity matching of multimedia query frames against reference content, reference database comprising of a cluster index using cluster keys to perform similarity matching and a multimedia index to perform sequence matching is built. Methods to update and maintain the reference database that enables addition and removal of the multimedia contents, including portions of multimedia content, from the reference database in a running system are described. Hierarchical multi-level partitioning methods to organize the reference database are presented. Smart partitioning of the reference multimedia content according to the nature of the multimedia content, and according to the popularity among the social media, that supports scalable fast multimedia identification is also presented. A caching mechanism for multimedia search queries in a centralized or in a decentralized distributed system and a client based local multimedia search system enabling multimedia tracking are described.