Gesture-Based Object Identification via Sensor Data Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current search techniques require users to accurately describe objects to identify them, which can be tedious and unsatisfactory, especially for objects without official names or difficult to describe, and do not efficiently allow users to acquire specific information about objects in their environment.

Innovation Solution

A method and system that utilize directional gestures and trigger events, such as voice prompts and sensor data analysis, to identify and characterize objects, enabling users to acquire information about objects using convenient multi-modal interactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If users provide accurate textual descriptions of objects to search engines, then search results can identify the object, but the process becomes tedious and time-consuming

Engineering Contradiction:
Improveobject identification accuracyVSAvoidtime required for object identification
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the mechanical process of manually typing textual descriptions with an optical system. A camera captures images of the object, and image recognition algorithms automatically identify the object, substituting the manual description process with automated visual analysis.

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

Solution Approach 2:

The system enables self-service object identification by allowing the camera to autonomously capture images and the algorithm to automatically recognize and identify objects without requiring user intervention for description or manual search queries.

Inventive Principle:
Principle #25Self-service

2Loss of information

If users manually describe physical attributes of objects, then the object can be identified, but additional searches are needed to locate specific information

Engineering Contradiction:
Improvecompleteness of object informationVSAvoidnumber of search steps required
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent merges multiple functions into a single integrated system: image capture, object identification, and information retrieval all occur in one automated process. The system combines camera imaging, pattern recognition, and database searching into a unified workflow that eliminates separate manual search steps.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system performs preliminary object identification and information retrieval automatically before the user needs the information. By pre-capturing images and pre-processing object data, the system prepares information in advance, eliminating the need for subsequent manual searches.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If users attempt to identify objects without accurate descriptions, then the process becomes unsatisfactory, but current techniques do not support objects without official names

Engineering Contradiction:
Improveability to identify various types of objectsVSAvoidaccuracy of object identification
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent changes the identification parameters from relying on textual descriptions and official names to using visual image parameters. By analyzing visual characteristics such as color, shape, texture, and patterns in images, the system can identify objects that lack formal names or descriptions, expanding adaptability while maintaining reliability through visual pattern recognition.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2891954B1User-directed personal information assistant
Publication Date: 2024.04.17 HARMAN INT IND INC
  • EP2891954B1 patent drawingFigure 1
  • EP2891954B1 patent drawingFigure 2A~2B
  • EP2891954B1 patent drawingFigure 3

AI summary

One embodiment of the present invention sets forth a technique for identifying a characteristic of an object. The technique includes determining (810) that the object is a target of a directional gesture and detecting (820) a trigger event that occurs while the object is the target of the directional gesture. The technique further includes, in response to the trigger event, acquiring (830) sensor data associated with the object via one or more sensors and analyzing (840) the sensor data to identify at least one characteristic of the object.