-
公开(公告)号：US11829523B2
公开(公告)日：2023-11-28
申请号：US17971600
申请日：2022-10-23
Applicant: BLINK TECHNOLOGIES INC.
Inventor： Gilad Drozdov , Oren Haimovitch-Yogev , Uri Wollner
IPC: G06F3/01 , G06T7/73 , G06V40/16 , G06V40/18 , G06T1/20 , G06T7/00 , A61B3/113 , G06V10/764 , G06V10/82 , G06V10/44 , G06V10/20
CPC classification number: G06F3/012 , A61B3/113 , G06F3/013 , G06T1/20 , G06T7/0012 , G06T7/73 , G06V10/255 , G06V10/44 , G06V10/764 , G06V10/82 , G06V40/161 , G06V40/171 , G06V40/18 , G06V40/193 , G06T2207/20084 , G06T2207/30041
Abstract: Anatomically-constrained gaze estimation providing a point of reference (PoR) in a 3D field and/or 2D plane, based on 6 DOF head pose constrained by the eyeball center being fixed in a common 3D coordinate system.
-
公开(公告)号：US11514720B2
公开(公告)日：2022-11-29
申请号：US16732640
申请日：2020-01-02
Applicant: BLINK TECHNOLOGIES INC.
Inventor： Oren Haimovitch-Yogev , Tsahi Mizrahi , Andrey Zhitnikov , Almog David , Artyom Borzin , Gilad Drozdov
IPC: G06V40/18 , G06K9/62 , G06N3/04 , G06N3/08 , G06T7/11 , G06V40/16 , G06V40/19 , G06V10/75 , G06V10/82
Abstract: The disclosure relates to systems, methods and programs for geometrically constrained, unsupervised training of convolutional autoencoders on unlabeled images for extracting eye landmarks. Disclosed systems for unsupervised deep learning of gaze estimation in eyes' image data are implementable in a computerized system. Disclosed methods include capturing an unlabeled image comprising the eye region of a user; and training a plurality of convolutional autoencoders on the unlabeled image comprising the eye region of a user using an initial geometrically regularized loss function to determine a plurality of eye landmarks.
-
公开(公告)号：US11513593B2
公开(公告)日：2022-11-29
申请号：US17298935
申请日：2019-11-28
Applicant: BLINK TECHNOLOGIES INC.
Inventor： Gilad Drozdov , Oren Haimovitch-Yogev , Uri Wollner
Abstract: The disclosure relates to systems, methods and programs for anatomically-constrained gaze estimation. More specifically, the disclosure is directed to systems, methods and programs for providing a point of reference (PoR) in a 3D field and/or 2D plane, based on 6 DOF head pose constrained by the eyeball center being fixed in a common 3D coordinate system.
-
公开(公告)号：US12141343B2
公开(公告)日：2024-11-12
申请号：US17551187
申请日：2021-12-15
Applicant: BLINK TECHNOLOGIES INC.
Inventor： Gilad Drozdov , Oren Haimovitch-Yogev , Tal Knafo
Abstract: The disclosure relates to systems, methods and programs implicitly determines the individual gaze correction required for each user when it comes to determining the point-of-regard and region of interest using a seamless-adaptation-process (SAP) in settings when calibration is not practical/desired (e.g. long-range display eye sensing). The present systems, methods, and programs also correct for variables arising from salient features such as user eye attributes (e.g., ocular dominance or chronic eyesight conditions) that are not visible within a digital frame, environmental conditions, and camera-screen settings. The method is not eye-tracking method dependent, can perform under all environmental conditions while the displayed content is known. The concept of a data driven seamless gaze correction can be extended beyond long-range displays to any gaze-enabled device where fixation accuracy can be improved (e.g., mobile phones, tablets etc.).
-
5.发明授权
公开(公告)号：US11776315B2
公开(公告)日：2023-10-03
申请号：US17518656
申请日：2021-11-04
Applicant: BLINK TECHNOLOGIES INC.
Inventor： Oren Haimovitch-Yogev , Gilad Drozdov , Igor Nor , Nadav Yehonatan Arbel
CPC classification number: G06V40/193 , G06F3/013 , G06V10/454 , G06V40/19
Abstract: The disclosure relates to systems, methods, and programs for implicitly determining the relation of the eyes associated with gaze inference, including salient features (e.g., ocular dominance) that are not visible within a digital frame, allowing for real-time determination of the user's dominant eye for the purpose of gaze estimation, and point-of-regard mapping onto a 2D plane, achieved through an end-to-end training of an eye selector agent with domain-expertise knowledge embedded in an unsupervised manner via a convolutional deep neural network training process.
-
公开(公告)号：US12093835B2
公开(公告)日：2024-09-17
申请号：US17971601
申请日：2022-10-23
Applicant: BLINK TECHNOLOGIES INC.
Inventor： Oren Haimovitch-Yogev , Tsahi Mizrahi , Andrey Zhitnikov , Almog David , Artyom Borzin , Gilad Drozdov
IPC: G06N3/088 , G06F18/214 , G06N3/04 , G06N3/0455 , G06N3/0464 , G06N3/08 , G06N20/10 , G06T7/11 , G06V10/75 , G06V10/764 , G06V10/82 , G06V40/16 , G06V40/18 , G06V40/19
CPC classification number: G06N3/088 , G06F18/2155 , G06N3/04 , G06N3/08 , G06T7/11 , G06V10/755 , G06V10/82 , G06V40/165 , G06V40/171 , G06V40/19 , G06V40/193 , G06T2207/20081 , G06T2207/20084 , G06T2207/20132
Abstract: Unsupervised, deep learning of eye-landmarks in a user-specific eyes' image data by capturing an unlabeled image comprising an eye region of a user, using an initial geometrically regularized loss function, training a plurality of convolutional autoencoders on the unlabeled image comprising the eye region of the user to recover a plurality of user-specific eye landmarks, training a convolutional neural network for autoencoded landmarks-based recovery from the unlabeled image, and where the initial geometrically regularized loss function is represented by the formula LAE=λreconLrecon+λconcLconc+λsepLsep+λeqvLeqv where LAE is total AutoEncoder Loss, λreconLrecon is λ-weighted reconstruction loss, λconcLconce is λ-weighted concentration loss, λsepLsep is λ-weighted separation loss, and λeqvLeqv is λ-weighted equivalence loss.
-
公开(公告)号：US12026980B2
公开(公告)日：2024-07-02
申请号：US17376388
申请日：2021-07-15
Applicant: BLINK TECHNOLOGIES INC.
Inventor： Gilad Drozdov , Nadav Arbel , Tsahi Mizrahi , Soliman Nasser , Artyom Borzin , Uri Wollner , Oren Haimovitch-Yogev
IPC: A61B3/00 , A61B3/11 , A61B3/113 , A61B3/14 , A61H5/00 , G02B27/00 , G02B27/01 , G06T7/00 , G06T7/13 , G06T7/73 , G06V40/19 , H04N13/302 , H04N13/327 , H04N13/383
CPC classification number: G06V40/19 , A61B3/0025 , A61B3/112 , G02B27/0093 , G02B27/017 , G06T7/13 , G06T2207/30041
Abstract: The disclosure relates to systems, methods and programs for developing real-time user-specific eye model based on iris localization using solely pupil-ellipse analysis.
