VR Caption Rendering Based on User Interest and Gaze

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing virtual reality systems fail to provide an immersive user experience due to synchronization issues between visual and audio content, lack of interaction with the physical environment, and limited freedom of movement, leading to a disjointed and less engaging experience.

Innovation Solution

Implementing a headset with integrated sensors and AI to track user movements and environmental interactions, synchronized audio and visual rendering, and dynamic captioning based on user interest, allowing 6DoF movement and augmented reality integration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If captions are always rendered for virtual reality objects, then information completeness is improved, but visual clutter and user distraction increase

Engineering Contradiction:
Improvecaption information completenessVSAvoidvisual distraction
Core Design Contradiction:
Loss of informationVSObject-affected harmful factors

Solution Approach 1:

The patent applies dynamics by making caption rendering conditional and adaptive rather than static. The system dynamically adjusts caption visibility based on real-time detection of user interest levels, determined through gaze tracking and interaction metrics. This resolves the contradiction by rendering captions only when and where users are actually interested, maintaining information completeness while eliminating unnecessary visual distraction.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system implements feedback mechanisms by continuously monitoring user behavior (gaze direction, focus duration, interaction patterns) and using this feedback to control caption rendering decisions. This closed-loop approach ensures captions are displayed based on actual user needs, balancing information delivery with visual comfort.

Inventive Principle:
Principle #23Feedback

2Manufacturing precision

If rendering quality is increased for all virtual reality content, then visual fidelity is improved, but computational load and processing time increase

Engineering Contradiction:
Improverendering qualityVSAvoidprocessing efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies local quality by differentiating rendering quality based on spatial location and user interest. High-quality rendering with captions is applied only to virtual reality objects within the user's field of view or periphery that have been identified as areas of interest. Objects outside the user's attention receive reduced or no caption rendering, optimizing the balance between visual fidelity and processing efficiency.

Inventive Principle:
Principle #3Local quality

3Reliability

If audio and visual content are synchronized perfectly, then immersion quality is improved, but system complexity increases

Engineering Contradiction:
Improvesynchronization accuracyVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system applies preliminary action by pre-synchronizing audio and visual content during the content creation and preparation phase. Timestamps and synchronization metadata are embedded in advance, allowing the playback system to maintain synchronization without requiring complex real-time coordination mechanisms, thus reducing system complexity while maintaining reliability.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3719613B1Rendering captions for media content
Publication Date: 2026.05.06 NOKIA TECHNOLOGIES OY
  • EP3719613B1 patent drawingFigure 1
  • EP3719613B1 patent drawingFigure 2
  • EP3719613B1 patent drawingFigure 3a

AI summary

An apparatus comprising means for rendering virtual reality content, wherein the virtual reality content comprises a virtual reality object associated with a caption and wherein the caption is rendered within a field of view of a user; determining a level of interest with respect to the virtual reality object; and determining if the caption is to be rendered based at least partly on the determined level of interest associated with the virtual reality object.