US 12,462,411 B2
Method and system for predicting distance of gazed objects using infrared (IR) camera
Rahul Dasharath Gavas, Bangalore (IN); Tince Varghese, Bangalore (IN); Ramesh Kumar Ramakrishnan, Bangalore (IN); Rolif Lima, Bangalore (IN); Priya Singh, Bangalore (IN); Shreyasi Datta, Bangalore (IN); Somnath Karmakar, Kolkata (IN); Mithun Basaralu Sheshachala, Bangalore (IN); and Arpan Pal, Kolkata (IN)
Assigned to TATA CONSULTANCY SERVICES LIMITED, Mumbai (IN)
Filed by Tata Consultancy Services Limited, Mumbai (IN)
Filed on May 7, 2024, as Appl. No. 18/656,863.
Claims priority of application No. 202321034600 (IN), filed on May 17, 2023.
Prior Publication US 2024/0386590 A1, Nov. 21, 2024
Int. Cl. G06T 7/55 (2017.01); G06T 7/73 (2017.01); G06V 10/30 (2022.01); G06V 10/32 (2022.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 40/18 (2022.01)
CPC G06T 7/55 (2017.01) [G06T 7/73 (2017.01); G06V 10/30 (2022.01); G06V 10/32 (2022.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 40/193 (2022.01); G06T 2207/10048 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30201 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A processor-implemented method for predicting distance of gazed objects, the method comprising:
pretraining a gaze predictor ML model via one or more hardware processor to predict distance of at least one gazed object positioned from eye of each subject during a systematic execution of a set of tasks;
receiving via the one or more hardware processors one or more IR images of each eye of each subject for fixed duration as input from a pair of IR cameras configured to either side of a spectacle;
acquiring via the one or more hardware processors one or more pupillary image information from each pupil of each eye from the one or more IR images;
extracting from each pupillary information of each pupil via the one or more hardware processors a set of features, and denoising eye blinks from the set of features, wherein the eye blinks from the set of features are denoised by performing the steps of:
obtaining the one or more pupillary image information from each pupil of each eye;
storing for each time window each pupillary image information in a temporary array;
identifying one or more eye blinks from the temporary array;
identifying a region of interest (ROI) by marking a sub window around each blink;
interpolating the ROI for each eyeblink based on (xL, yL) which is leftmost data point in the ROI and (xR, yR) is the rightmost point in the ROI; and
applying a smoothening filter over each pupillary image information in the temporary array to remove sharp points from each eye blink;
predicting via the one or more hardware processors a distance of a gazed object from current location of the subject using the gaze predictor ML model and the set of features; and
classifying via the one or more hardware processors the gazed object of the subject based on the distance into at least one of a near class, an intermediate class, and a far class.