Region-of-interest imaging and identifying eye features
AssigneeMeta Platforms Technologies, LLC
InventorsQing Chao, Robert Dale Cavin, Guillaume Lestoquoy, Mohammadhossein Daraeihajitooei, Xinqiao Liu, Raffaele Capoccia, Sascha Hallstein, Ziyun Li
Life of this patent
Estimating the term…
Abstract
Imaging signals are generated in response to image light. Event signals are generated in response to receive the imaging signals from imaging pixels. A Region of Interest (ROI) of the imaging pixels is identified from a spatial concentration of event signals in the ROI of imaging pixels within a time period. An ROI portion of the imaging pixels in the ROI are driven to capture an ROI image frame.
Claims20
An imaging sensor comprising:
- imaging pixels configured to generate imaging signals in response to image light;
- event-sensing logic configured to generate event signals in response to receiving the imaging signals from the imaging pixels;
- and
- region-of-interest (ROI) logic coupled to receive the event signals from the event-sensing logic, wherein the ROI logic is configured to:
- identify an ROI of the imaging pixels from a spatial concentration of event signals in the ROI within a time period;
- and
- drive an ROI portion of the imaging pixels in the ROI to capture an ROI image frame.
A computer-implemented method comprising:
- generating imaging signals in response to image light;
- generating event signals in response to receive the imaging signals from imaging pixels;
- identifying a Region of Interest (ROI) of the imaging pixels from a spatial concentration of event signals in the ROI of imaging pixels within a time period;
- and
- driving an ROI portion of the imaging pixels in the ROI to capture an ROI image frame.
A computer-implemented method comprising:
- illuminating an eyebox region with a fringe illumination pattern;
- capturing a first image of the eyebox region at a first time period while the eyebox region is illuminated with the fringe illumination pattern;
- capturing a second image of the eyebox region at a second time period while the eyebox region is illuminated with the fringe illumination pattern;
- generating intensity difference data between the first image and the second image;
- and
- identifying an eye feature based at least in part on the intensity difference data.
Dependencies are inferred from the claim text.