Smart Home Gesture Control With Two-Stage Motion Verification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional terminal devices experience a high false drop rate in gesture recognition due to unintentional user gestures, leading to inaccurate control operations.

Innovation Solution

A method and system that determines a to-be-recognized region in an environmental image, monitors target body part movements, and controls the terminal device based on consistent activation and control movements, reducing false drops by using a two-stage movement verification process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If gesture recognition is implemented using continuous image collection and hand movement tracking, then the terminal device can perform function operations based on user gestures, but the false drop rate increases due to unintentional gestures being misrecognized

Engineering Contradiction:
Improvegesture-based control automationVSAvoidfalse drop rate
Core Design Contradiction:
Extent of automationVSReliability

Solution Approach 1:

The gesture recognition process is segmented into distinct stages: initial gesture detection, confirmation phase, and execution phase. The to-be-recognized region is further segmented into sub-regions for detailed movement analysis. This segmentation allows the system to verify gestures through multiple checkpoints, reducing false recognition of unintentional movements while maintaining automation.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If the terminal device monitors and recognizes all hand movements in the environment, then gesture control functionality is enhanced, but the probability of false drops increases due to unintentional gestures

Engineering Contradiction:
Improvegesture control functionalityVSAvoidfalse drop probability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

Instead of uniformly monitoring all regions, the system identifies and focuses on a specific to-be-recognized region where gestures are most likely to occur. Within this region, monitoring intensity and recognition sensitivity are optimized. This localized approach enhances gesture control functionality in relevant areas while reducing false drops from movements in unrelated regions.

Inventive Principle:
Principle #3Local quality

3Speed

If the system uses a single-stage gesture recognition process, then the response speed is fast, but the accuracy of distinguishing intentional from unintentional gestures is low

Engineering Contradiction:
Improvegesture response speedVSAvoidgesture intention recognition accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The system performs preliminary actions before final gesture execution: first detecting potential gestures in a to-be-recognized region, then verifying them through additional monitoring stages. This preliminary detection and verification process ensures accurate distinction between intentional and unintentional gestures while maintaining acceptable response speed through optimized processing priorities.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4105765B1Device control method, apparatus and system
Publication Date: 2025.09.10 HUAWEI TECH CO LTD
  • EP4105765B1 patent drawingFigure 1~2
  • EP4105765B1 patent drawingFigure 3~4
  • EP4105765B1 patent drawingFigure 5~6

AI summary

A device control method, apparatus, and system applicable to the field of smart home technologies. The method includes: recognizing a collected environmental image; determining at least one to-be-recognized region in the environmental image (201); monitoring a target body part movement in the to-be-recognized region (202); if it is monitored that the target body part movement is consistent with a preset activation movement, tracking and monitoring a target body part movement in a target to-be-recognized region (203); or if it is monitored that the target body part movement in the target to-be-recognized region is a preset control movement, controlling a terminal device to perform a function operation corresponding to the control movement (204). The terminal device (120) is controlled by using a target body part movement including two-stage movements. This reduces a probability that a user controls the terminal device (120) by triggering a target body part movement once, reduces a false drop rate of monitoring the triggered movement by an image collection device (110), and improves accuracy of controlling the terminal device (120) by the image collection device (110) based on the movement triggered by the user.