3D Face Mesh Expansion for Personalized AR Graphical Elements

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing digital platforms limit users' ability to customize content items, such as emojis, to accurately resemble their physical features, and these customizations are often static and lack interactive capabilities, discouraging creative expression and engagement.

Innovation Solution

A social networking system generates customized graphical elements by creating a 3D face mesh from a user's image, which can be expanded to include recognizable features beyond the face, and applies AR filters to create dynamic and interactive content items that reflect individual users' unique physical qualities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If users are provided with limited customization options for content items, then the system is easier to operate and manage, but users cannot accurately represent their unique physical features

Engineering Contradiction:
Improveaccuracy of physical feature representationVSAvoidcustomization system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system creates a digital 3D copy of the user's face by generating a face mesh from uploaded images. This mesh accurately replicates the user's physical features including facial geometry, skin texture, and anatomical structures, allowing precise representation without manual customization of individual features

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The system transforms static 2D images into dynamic 3D models by adjusting dimensional parameters. The face mesh evolves from fixed image data into an interactive three-dimensional representation that can be viewed from multiple angles and used across different digital platforms

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If content items are made static, then they are simpler to implement and store, but they lack interactive capabilities and engagement

Engineering Contradiction:
Improveinteractive capabilityVSAvoidcontent item complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent transforms static emoji images into dynamic 3D models that can respond to user interactions, device movements, and environmental contexts. The face mesh can rotate, scale, and adapt to different viewing conditions, enabling interactive capabilities while maintaining the core emoji functionality

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The 3D face mesh serves multiple functions across different platforms and applications. The same mesh can be used for profile pictures, stickers, filters, and interactive elements in various digital environments, eliminating the need for separate customizations for each platform

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If extensive customization options are provided, then users can accurately represent their physical features, but the creation process becomes labor intensive

Engineering Contradiction:
Improveaccuracy of physical feature representationVSAvoidcontent creation efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system performs automatic face detection, mesh generation, and feature extraction without requiring user intervention. The uploaded images are automatically processed to create the 3D face mesh, which then serves as the basis for all customizations, eliminating manual measurement and adjustment work

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system pre-generates the complete 3D face mesh from uploaded images before any customization is needed. This preliminary 3D model contains all facial features and geometries ready for use, so when users want to create customized content, the foundation is already prepared and can be quickly applied

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12505494B2Generating customized graphical elements from user-provided images
Publication Date: 2025.12.23 META PLATFORMS INC
  • US12505494B2 patent drawing
  • US12505494B2 patent drawing
  • US12505494B2 patent drawing

AI summary

Techniques for generating customized graphical elements are described. For example, a social networking system may receive, from a device of an account of a social networking system, an image including at least a portion of a face of a user associated with the account. The social networking system may generate a face mesh including at least a portion of the face depicted in the image, and then may expand the face mesh to include at least a portion of the head of the user other than the face to generate a customized graphical element. The customized graphical element may be further modified, stored, and subsequently shared with one or more other accounts alone and/or in association with one or more content items.