Sensor Agnostic Gesture Detection via Threshold Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing IoT systems face challenges in usability due to device-specific gestures, which become ineffective if the user forgets or is unable to use the device originally associated with the gesture, limiting the flexibility and accessibility of gesture-based interactions across different IoT devices and environments.
Innovation Solution
A ubiquitous gesture agent is used to record and convert gestures from one type of sensor output into a format measurable by another type of sensor, allowing gestures to be recognized and executed across various devices and environments, including smart buildings, without the need for the original device, utilizing machine learning algorithms for data transformation and threshold generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If device-specific gestures are used in IoT systems, then gesture recognition accuracy is improved, but usability and accessibility deteriorate when users forget or cannot use the original device
Solution Approach 1:
The patent implements a ubiquitous gesture agent that stores gesture data and transforms triggering thresholds across different sensor types, enabling gestures recorded on one device to be recognized and executed on multiple different devices. This makes the gesture control system universal rather than device-specific, allowing users to trigger actions on any associated device regardless of which device they use to perform the gesture.
Solution Approach 2:
The ubiquitous gesture agent serves as an intermediary between the user's gesture input and the target IoT devices. It receives gesture data from one device, transforms the triggering thresholds to match different sensor types, and delivers the transformed data to appropriate devices, thereby mediating the interaction and eliminating the need for users to remember device-specific gestures.
2Measurement precision
If gestures are tied to specific devices, then sensor output precision is improved, but system flexibility and adaptability deteriorate
Solution Approach 1:
The patent transforms triggering thresholds from one sensor type to another by changing the parameters to match the characteristics of different sensors. The system adjusts the threshold values and data formats to accommodate various sensor types (e.g., accelerometer, gyroscope, camera) while preserving the essential gesture characteristics, thereby enabling cross-device gesture recognition without sacrificing precision.
Solution Approach 2:
The ubiquitous gesture agent enables a single gesture recording to be adapted for use with multiple different sensor types and devices. By storing the gesture data and transforming it for different sensor formats, the system makes the gesture control mechanism universal, allowing the same gesture to work across diverse IoT devices with different sensor capabilities.
3Adaptability or versatility
If multiple device-specific gesture systems are maintained, then device functionality is improved, but system complexity increases
Solution Approach 1:
The patent extracts the gesture data storage and transformation functionality into a separate ubiquitous gesture agent, independent of any specific device. This centralizes the complex operations of gesture recognition, threshold transformation, and cross-device coordination in one dedicated component, thereby reducing the complexity burden on individual IoT devices while maintaining comprehensive functionality.
Solution Approach 2:
The system merges multiple device-specific gesture handling capabilities into a single ubiquitous gesture agent that manages all gesture data. By combining the functions of recording, storing, transforming, and distributing gesture information in one centralized service, the system reduces overall complexity while maintaining the ability to support multiple devices and sensor types.
Data Source
AI summary
Described are techniques for sensor agnostic gesture detection. The techniques include recording a gesture using sensor output from a first user device to a first user profile associated with a ubiquitous gesture agent. The techniques further include associating the gesture with a triggering threshold indicating the gesture and based on the sensor output, and a processor-executable action that is executable by one of a plurality of user devices associated with the user profile. The techniques further include transforming the triggering threshold into a corresponding triggering threshold for a different type of sensor. The techniques further include identifying the gesture using data from the different type of sensor that satisfies the corresponding triggering threshold. The techniques further include implementing, by the one of the plurality of user devices, the processor-executable action associated with the gesture.