Multi-Device Gesture Onset Detection Using Segmented Sensor Activation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurate detection of gestures using unreliable sensors like accelerometers and gyroscopes is challenging, especially when determining the direction of a device relative to another device, as it requires additional verification and often relies on costly, power-consuming sensors that are continuously active.
Innovation Solution
A multi-step gesture detection technique that uses a gesture onset detector on one device to trigger additional sensors on other devices only when necessary, combining data from multiple devices, including inertial measurement units, cameras, and microphones, to improve accuracy and reduce power consumption and privacy risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If costly and power-consuming sensors are continuously active to improve gesture detection accuracy, then measurement precision is improved, but use of energy increases
Solution Approach 1:
The patent implements periodic action by having sensors transition between active and inactive states. Economical sensors continuously monitor for gesture onsets, while costly precision sensors are activated only periodically when a gesture is detected, maintaining high detection accuracy while reducing overall power consumption.
Solution Approach 2:
The patent applies preliminary action by using economical sensors to detect gesture onsets before activating costly precision sensors. This preliminary detection stage filters out non-gesture movements, ensuring that expensive sensors are only activated when truly needed, thus balancing accuracy and power consumption.
2Reliability
If multiple sensors are continuously active to verify gesture direction, then reliability is improved, but use of energy increases
Solution Approach 1:
The system uses periodic action by activating multiple sensors only when a gesture onset is detected. Economical sensors continuously monitor and trigger the costly precision sensors only at relevant moments, maintaining reliable gesture verification while minimizing continuous power consumption.
Solution Approach 2:
Preliminary action is implemented through a two-stage detection process where economical sensors first detect potential gestures, then trigger the activation of costly precision sensors for verification. This ensures reliable gesture direction verification only when necessary.
3Measurement precision
If costly sensors are used to detect gesture direction, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the sensor system into two functional segments: economical sensors for continuous gesture onset detection and costly precision sensors for directional verification. This segmentation allows the system to achieve high measurement precision only when needed, reducing overall device complexity.
Solution Approach 2:
Preliminary action is used to filter gesture candidates using economical sensors before engaging costly precision sensors. This preliminary filtering reduces the frequency and complexity of operations involving expensive sensors, simplifying the overall device architecture.
4Use of energy by moving object
If economical sensors are used for initial detection, then use of energy is reduced, but measurement precision deteriorates
Solution Approach 1:
The system segments detection into two stages: initial detection using economical sensors with lower power consumption, and verification using costly precision sensors. This segmentation maintains overall measurement precision while significantly reducing average power consumption.
Solution Approach 2:
Periodic action is implemented by having economical sensors continuously operate at low power and triggering costly precision sensors only periodically when gestures are detected, achieving a balance between continuous monitoring capability and energy efficiency.
Data Source
AI summary
Implementations of the disclosed subject matter provide techniques for improved identification of a gesture based on data obtained from multiple devices. A method may include receiving an indication of an onset of a gesture, from a first device, at a gesture coordinating device. Next, first subsequent data describing the gesture may be received from a second device, at the gesture coordinating device. Based on the indication and the first subsequent data, the gesture may be identified. In response to identification of the gesture, an action may be performed based on the gesture identified. In some cases, the gesture coordinating device may be a cloud-based device.