Media Item Descriptor Matching for Design Template Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing digital design software applications are ineffective in identifying the most suitable design templates from a plethora of options, making it challenging for users to select the appropriate template for their designs, especially when hundreds of thousands of templates are available.

Innovation Solution

A computer-implemented method that utilizes a machine learning model to analyze input media items and generate descriptors, which are then compared to descriptors of stored design templates to identify and display suitable candidate templates, allowing users to preview and replace existing media items with their own input media items.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a large number of design templates are provided to users, then the variety and quality of design options are improved, but the difficulty of selecting the appropriate template increases

Engineering Contradiction:
Improvedesign template varietyVSAvoidtemplate selection difficulty
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system automatically analyzes the input media item and retrieves matching design templates without requiring user manual search. The machine learning model processes the media item characteristics and autonomously identifies suitable templates, eliminating the need for users to manually browse or search through hundreds of thousands of templates.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

A machine learning model acts as an intermediary between the user's input media item and the design templates. The model analyzes the media item characteristics and mediates the matching process by comparing features such as color, style, and composition to retrieve the most relevant templates, simplifying the selection process.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If manual search methods are used to find design templates, then user control over the search process is maintained, but time consumption and efficiency are reduced

Engineering Contradiction:
Improveuser controlVSAvoidtemplate search time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The manual mechanical search process is replaced with an automated machine learning-based retrieval system. Instead of users manually browsing or using search queries, the system uses AI algorithms to automatically analyze media items and retrieve matching templates, significantly reducing search time while maintaining relevance.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary analysis of the input media item before template retrieval. The machine learning model pre-processes the media item to extract characteristics and features, which then guides the template matching process, ensuring efficient and accurate results without requiring user intervention.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If design templates are manually browsed and searched, then users can filter by specific criteria, but the complexity of the search interface increases

Engineering Contradiction:
Improvesearch criteria filteringVSAvoidsearch interface complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system automatically determines and applies relevant filtering criteria based on the input media item characteristics. The machine learning model analyzes the media item and autonomously identifies appropriate template categories, styles, and features, eliminating the need for users to manually configure complex search filters or interfaces.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4428743A1Systems and methods for identifying a design template matching a media item
Publication Date: 2024.09.11 CANVA PTY LTD
  • EP4428743A1 patent drawingFigure 1
  • EP4428743A1 patent drawingFigure 2
  • EP4428743A1 patent drawingFigure 3

AI summary

A method for automatically generating one or more digital designs is disclosed. The method includes identifying an input media item; processing the input media item to generate an input media item descriptor; and identifying a first target media item from a set of target media items. Each target media item in the set of target media items is associated with a target media item descriptor and a candidate design template, and the first target media item is identified based on a similarity between the input media item descriptor and the target media item descriptor of the first target media item. The method further includes generating a new digital design. The new digital design being based on the candidate design template associated with the first target media item, and generated to replace the first target media item with the input media item.