US 12,287,915 B2
Camera system for pupil detection and eye tracking
Kevin Boyle, San Francisco, CA (US); Robert Konrad, San Francisco, CA (US); and Nitish Padmanaban, Menlo Park, CA (US)
Assigned to Sesame AI, Inc., San Francisco, CA (US)
Filed by Sesame AI, Inc., San Francisco, CA (US)
Filed on Nov. 15, 2022, as Appl. No. 17/987,444.
Application 17/987,444 is a continuation of application No. 17/552,741, filed on Dec. 16, 2021, granted, now 11,543,883.
Claims priority of provisional application 63/126,750, filed on Dec. 17, 2020.
Prior Publication US 2024/0160012 A1, May 16, 2024
Int. Cl. G06F 3/01 (2006.01); G02B 27/00 (2006.01); G02B 27/01 (2006.01); G06T 7/70 (2017.01); G06V 40/18 (2022.01); H04N 23/71 (2023.01); H04N 23/74 (2023.01)
CPC G06F 3/013 (2013.01) [G02B 27/0093 (2013.01); G06T 7/70 (2017.01); G06V 40/193 (2022.01); H04N 23/71 (2023.01); H04N 23/74 (2023.01); G02B 27/0172 (2013.01); G06T 2207/10048 (2013.01); G06T 2207/10152 (2013.01); G06T 2207/30201 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An eye tracking system comprising:
a first light source that emits a first pulse of light over a first time period, the first pulse of light directed along an optical path towards an eye of a user, a surface of the eye reflecting a portion of the first pulse of light back as reflected light along the optical path at a first brightness towards a target area;
a camera located in the target area, the camera and the first light source aligned with the optical path, the camera including a plurality of sensing pixels, each sensing pixel of a subset of the plurality of sensing pixels configured to:
detect an intensity value corresponding to a portion of the reflected light, and
responsive to the detected intensity value above a threshold level, output a data value representing a difference between a previous intensity value detected by each sensing pixel of the subset of sensing pixels and the detected intensity value; and
a controller configured to:
group data values output by the subset of sensing pixels into a two-dimensional matrix,
identify a pupil of the eye based on the data values grouped into the two-dimensional matrix, and
determine a gaze location for the eye based at least in part on the identified pupil.