Distributed Gesture Disambiguation via Client-Server Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current gesture recognition systems are resource-intensive and can lead to delayed user experiences due to the need for local processing and potential server unavailability, limiting the complexity and speed of gesture recognition, especially in mobility situations.

Innovation Solution

The system optimizes gesture recognition by progressively refining pattern matches using server-based computational resources, allowing incremental candidate matches to be provided to the client device, reducing computational load and enabling faster identification through intelligent processing and bandwidth management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If gesture recognition is performed locally on the device, then gesture identification can proceed without server availability, but computational resources and power consumption increase significantly

Engineering Contradiction:
Improvegesture identification reliabilityVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The gesture recognition system is segmented into two parts: a lightweight local component that performs initial gesture capture and basic filtering, and a remote server component that performs complex pattern matching and identification. This segmentation allows the device to maintain reliability while reducing local computational burden and power consumption.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A server acts as an intermediary between the gesture-capturing device and the final identification process. The server receives gesture data, performs resource-intensive pattern matching, and returns identification results, thereby relieving the client device of heavy computational tasks while maintaining reliable gesture recognition.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If complete gesture data is transmitted to the server, then accurate gesture identification can be achieved, but transmission time and network bandwidth consumption increase

Engineering Contradiction:
Improvegesture pattern matching accuracyVSAvoidgesture identification time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions on the client side by capturing and pre-processing gesture data before transmission. Basic filtering and initial analysis are conducted locally to reduce the volume of data that needs to be transmitted, thereby maintaining identification accuracy while reducing transmission time and network bandwidth consumption.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If a comprehensive gesture library is stored locally, then gesture recognition can proceed offline, but memory requirements and device complexity increase

Engineering Contradiction:
Improveoffline gesture recognition capabilityVSAvoidlocal processing complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The comprehensive gesture library and complex pattern matching algorithms are extracted from the client device and relocated to a remote server. This extraction allows the device to maintain offline capability for basic gesture capture while avoiding the complexity of storing extensive gesture libraries and running sophisticated recognition algorithms locally.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP2818978B1System and method for gesture disambiguation
Publication Date: 2023.08.30 ORANGE SA
  • EP2818978B1 patent drawingFigure 1~2
  • EP2818978B1 patent drawingFigure 3
  • EP2818978B1 patent drawingFigure 4~5

AI summary

Optimized distributed disambiguation of gestures is disclosed. In one innovative aspect, there is a gesture controlled device. The gesture controlled device includes a sensor configured to detect an input gesture. The gesture controlled device also includes a gesture detector configured to obtain the input gesture. The gesture controlled device further includes a transmitter configured to transmit a candidate gesture request including a first portion of the input gesture, the transmission based at least in part on a resource characteristic of the gesture controlled device. The gesture controlled device also includes a receiver configured to receive a set of candidate gestures for the candidate gesture request. The gesture detector is configured to identify the input gesture based on a second portion of the input gesture and the received set of candidate gestures.