Dynamic 3D Avatars for Personalized Instant Messaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional electronic messaging lacks personalization, particularly in instant messaging, as generic emojis and graphics fail to replicate the intimacy and emotional nuances of in-person communication.

Innovation Solution

A method and system for creating customized animatable 3D models of virtual characters that mimic the facial expressions of users, using input from audio and visual streams to generate dynamic animations, allowing for personalized and expressive communication through the creation of selectable links for electronic messages.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If generic emojis and graphics are used in electronic messaging, then the messaging can be transmitted efficiently, but the personalization and emotional connection are lost

Engineering Contradiction:
ImprovepersonalizationVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system creates a virtual 3D avatar that copies and replicates the user's facial expressions, gestures, and body language in real-time during communication. This copy approach allows the avatar to represent the user's emotional state and communication style, providing personalization without requiring complex custom content creation for each message.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent implements dynamic 3D avatars that can change their appearance, expressions, and animations in real-time based on the ongoing conversation. The avatars dynamically adjust their facial expressions and body language to match the user's current emotional state and communication context, enabling adaptive personalization as the conversation evolves.

Inventive Principle:
Principle #15Dynamics

2Adaptability or versatility

If customized 3D avatars with real-time animation are implemented, then personalization and emotional expression are improved, but the processing time and computational resources increase

Engineering Contradiction:
Improveemotional expressionVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system replaces complex real-time 3D rendering and animation calculations with pre-computed motion capture data and template-based animations. Instead of calculating every facial muscle movement and body gesture in real-time, the system substitutes these complex mechanical computations with pre-recorded motion sequences that can be quickly applied to the avatar, significantly reducing processing time while maintaining expressive capability.

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

3Measurement precision

If real-time facial expression tracking and 3D model generation are performed, then the accuracy of emotion representation is improved, but the device resources and energy consumption increase

Engineering Contradiction:
Improveemotion representation accuracyVSAvoidenergy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system extracts and focuses on specific key facial features and body language cues that are most indicative of emotional state, rather than processing every pixel and muscle movement. By taking out only the essential expressive elements (such as eye movements, mouth shape, head orientation), the system achieves accurate emotion representation with significantly reduced computational energy consumption.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS11062494B2Electronic messaging utilizing animatable 3D models
Publication Date: 2021.07.13 DIDIMO INC
  • US11062494B2 patent drawing
  • US11062494B2 patent drawing
  • US11062494B2 patent drawing

AI summary

Dynamically customized animatable 3D models of virtual characters (“avatars”) in electronic messaging are provided. Users of instant messaging are represented dynamically by customized animatable 3D models of a corresponding virtual character. An example method comprises receiving input from a mobile device user, the input being an audio stream and/or an image/video stream; and based on an animatable 3D model and the streams, automatically generating a dynamically customized animatable 3D model corresponding to the user, including performing dynamic conversion of the input into an expression stream and corresponding time information. The example method includes generating a link to the expression stream and corresponding time information, for transmission in an instant message, and causing display of the customized animatable 3D model. Link generation and causing display is performed automatically or in response to user action. The animatable 3D model can be customized in the cloud or downloaded for customization.