Virtual Reality User Image Generation From Expression Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users wearing virtual reality (VR) headsets often lack a webcam for video conferencing and are reluctant to appear on calls with their headset visible, leading to the need for efficient avatar solutions that are not universally compatible across different communication platforms.

Innovation Solution

A software-based method generates and renders a user's image on a 2D display using expression tracking and a generative model, allowing for photorealistic avatars without a full 3D animation pipeline, and can be integrated with various communication systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If a full 3D animation pipeline is used to generate photorealistic avatars, then the visual realism is improved, but the computational cost and system complexity increase

Engineering Contradiction:
Improvevisual realismVSAvoidsystem complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential components needed for photorealistic avatar generation from the full 3D animation pipeline. Instead of implementing complete 3D modeling, rigging, and animation systems, the invention uses a generative model that directly synthesizes photorealistic images from input data, eliminating unnecessary pipeline complexity while maintaining visual quality

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the mechanical 3D animation pipeline with a software-based generative model. Instead of using traditional computer graphics rendering engines and 3D models, the system employs machine learning-based image generation that directly produces photorealistic avatars, substituting complex mechanical processing with intelligent software synthesis

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

2Manufacturing precision

If a full 3D animation pipeline is implemented, then photorealistic avatars can be generated, but compatibility across different communication platforms decreases

Engineering Contradiction:
ImprovephotorealismVSAvoidplatform compatibility
Core Design Contradiction:
Manufacturing precisionVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal avatar generation system that produces output compatible with multiple communication platforms. The generative model generates standardized image formats and protocols that can be integrated with various video conferencing systems (Zoom, Teams, etc.), making the photorealistic avatar technology platform-agnostic and broadly applicable

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

Solution Approach 2:

The patent adjusts key parameters of the generative model to optimize for both photorealism and platform compatibility. By tuning parameters such as image resolution, frame rate, and output format, the system achieves high visual quality while ensuring broad compatibility across different communication platforms and devices

Inventive Principle:
Principle #35Parameter changes

3Productivity

If expression tracking and generative models are used instead of full 3D pipelines, then computational cost is reduced, but the challenge remains of generating realistic expressions without 3D data

Engineering Contradiction:
Improvecomputational efficiencyVSAvoidexpression realism
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent introduces expression tracking data as an intermediary between the input video stream and the generative model. The tracking system extracts facial landmark positions and expression parameters, which serve as conditional inputs to guide the generative model in producing realistic facial expressions that match the user's actual emotions, bridging the gap between 2D processing and 3D-like realism

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4622248A1Virtual reality user image generation
Publication Date: 2025.09.24 CMS CAMERON MCKENNA NABARRO OLSWANG LLP
  • EP4622248A1 patent drawingFigure 1
  • EP4622248A1 patent drawingFigure 2A
  • EP4622248A1 patent drawingFigure 2B

AI summary

Images are generated and rendered on a first device having a two-dimensional display. The first device receives from a second device expression data indicative of a current facial expression of a user of the second device, where the second device has a three-dimensional display. The expression data is input to a generative model trained on an enrollment image indicative of a baseline image of the user's face. Facial image information is received from the generative model that is usable to render a two-dimensional image of the current facial expression on the first device. The facial image information is sent to the first device for rendering of the two-dimensional image of the current facial expression on the two-dimensional display in context of an on-going session of a communications system.