Extended Reality Interaction Mode Switching by Action Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing extended reality (XR) devices require manual selection of interaction modes, limiting efficiency and user experience as they cannot automatically distinguish between handle tracking and gesture recognition.

Innovation Solution

An extended reality interaction method that automatically determines user actions as handle or gesture actions through data acquisition, using reinforcement learning and template matching, to trigger corresponding interaction modes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If manual selection of interaction mode is implemented, then the system can support both handle tracking and gesture recognition, but the interaction efficiency deteriorates due to requiring user configuration

Engineering Contradiction:
Improveinteraction mode supportVSAvoidinteraction efficiency
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The system automatically detects user actions and switches interaction modes without requiring manual user configuration. The XR device monitors action data, determines action types, and autonomously triggers the appropriate interaction mode (handle tracking or bare hand tracking), making the system self-adapting to user needs.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The interaction mode is made dynamic and adaptable rather than static. The system can switch between handle tracking mode and bare hand tracking mode based on real-time action recognition, allowing the interaction mechanism to evolve according to user behavior rather than being fixed in advance.

Inventive Principle:
Principle #15Dynamics

2Device complexity

If fixed interaction mode is selected, then the system configuration is simplified, but the ease of operation deteriorates as users cannot switch between handle and gesture actions

Engineering Contradiction:
Improvesystem configurationVSAvoidaction flexibility
Core Design Contradiction:
Device complexityVSEase of operation

Solution Approach 1:

The XR device is designed to support multiple interaction modes (handle tracking and bare hand tracking) within a single system. The action recognition system can identify both types of actions and trigger corresponding interaction modes, making the device universal and adaptable to different user preferences and scenarios without requiring separate configurations.

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

3Productivity

If automatic action recognition is implemented, then interaction efficiency is improved, but the device complexity increases due to reinforcement learning and template matching systems

Engineering Contradiction:
Improveinteraction efficiencyVSAvoidrecognition system
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The action recognition system is divided into distinct modules: action data acquisition, action type determination (using reinforcement learning and template matching), and interaction mode triggering. This segmentation allows each component to be optimized independently and facilitates easier implementation and maintenance of the complex recognition system.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS12449911B2Extended reality interaction method and apparatus, electronic device and storage medium
Publication Date: 2025.10.21 BEIJING ZITIAO NETWORK TECH CO LTD
  • US12449911B2 patent drawing
  • US12449911B2 patent drawing

AI summary

The present disclosure provides an extended reality interaction method and apparatus, an electronic device and a storage medium, the extended reality interaction method includes: acquiring action data of a user in an extended reality scene; determining an action type of the user according to the action data; and triggering a corresponding interaction mode according to the action type, the corresponding interaction mode comprising a handle tracking mode and a bare hand tracking mode, according to the method provided by the present disclosure, an action type of the user is recognized and then a target interaction mode according to the action type is triggered for computing processing, without the need to manually set the interaction mode in advance.