Eye Tracker System for Diagnostics (CIDER)
Deepak Ganesan, Ph.D.
This invention is a highly-optimized, low-power wearable eye tracker system for detecting eye parameters including eye movement, pupil center, pupil diameter (i.e., dilation), blink duration, and blink frequency. The eye parameters may then be used to determine a variety of physiological and psychological conditions in humans. The system operates at a ten-fold reduction in power usage as compared to current eye tracker systems and methods, may be integrated into wearable, lightweight glasses, and does not require active calibration by the user. The eye tracker device includes an imaging component, an illuminator component, a photodiode component and a controller. The system incorporates a neural network enabling trade-offs between power consumption and robustness to illumination conditions, as well as between sensing and computational modes. The system can operate at very high frame rates (exceeding 100 fps) during typical operation.
• Low power, 10x less than state-of-the-art • Small form factor, integrable into eyeglasses • Lightweight • Orders of magnitude cheaper than state-of-the-art: $100 vs. $10,000
• Disease diagnosis • Athletic performance monitoring • Visual marketing • Safety monitoring • Technical training • VR/AR
Deepak Ganesan is a Professor in the Department of Computer Science at UMass Amherst. His research focuses on ultra-low power wireless communication via backscatter, novel platforms and algorithms for mobile and wearable health sensing, learning and inference on multi-modal sensor data, and micro-powered sensors. Dr. Ganesan leads the UMass Sensors Research Group.
Available for Licensing and/or Sponsored Research
UMA 16-014
U.S. Patent 10,016,130 Issued
|
