Information Technology
Imaging Technique for Recognizing Hand Gestures & Other Micromotions in 3-D
WARF: P170202US01
Inventors: Mohit Gupta, Brandon Smith, Pratham Desai, Vishal Agarwal
The Wisconsin Alumni Research Foundation (WARF) is seeking commercial partners interested developing a cost-competitive speckle imaging technique for measuring non-rigid micromotions on smartphones and other devices.
Overview
With recent advancements in digital imaging, image sensors have become popular for capturing motion in 3-D. However, measuring small scale motion such as fine hand gestures across a user interface remains challenging. Conventional low-cost cameras can detect motions only down to the millimeter or centimeter scale.
Speckle imaging is a technique currently used to measure microscale motions in a variety of settings, including industrial inspection, scientific imaging and user interfaces (e.g., optical mice). To date, speckle imaging has been limited to measuring 2-D motions of single rigid objects and not suitable for tracking motion toward or away from the sensor (i.e., axial motion).
Speckle imaging is a technique currently used to measure microscale motions in a variety of settings, including industrial inspection, scientific imaging and user interfaces (e.g., optical mice). To date, speckle imaging has been limited to measuring 2-D motions of single rigid objects and not suitable for tracking motion toward or away from the sensor (i.e., axial motion).
The Invention
UW–Madison researchers have developed a new imaging technique that analyzes speckle patterns to track extremely small 3-D motions on the order of 10-100 microns. This technique enables, for the first time, precise 3-D measurement of multiple moving objects using low-cost, off-the-shelf components.
Applications
- Gesture recognition for devices such as smartphones and wearable sensors
- Motion analysis of cells, molecules and other microscopic particles
Key Benefits
- Measures 3-D micromotions of more than one object
- Motion sensitivity is one to two orders of magnitude better than conventional sensors.
- Uses inexpensive components that can be scaled down
Stage of Development
Feasibility has been demonstrated with a hardware prototype consisting of a lens-less sensor and a laser pointer. The researchers have developed a high speed and accurate finger gesture recognition system that can be incorporated into various interactive wearable devices.
Additional Information
For More Information About the Inventors
Tech Fields
For current licensing status, please contact Michael Carey at [javascript protected email address] or 608-960-9867