-
公开(公告)号：US11664089B2
公开(公告)日：2023-05-30
申请号：US16575277
申请日：2019-09-18
申请人： SHARECARE AI, INC.
IPC分类号： G06F16/00 , G16B30/00 , G16B40/00 , G16B10/00 , G06F16/22 , G16B20/20 , G16B45/00 , G16B20/40 , G16B5/00 , G16B40/10 , C12Q1/6827 , C12Q1/6869 , G06F17/18 , G16B50/00
CPC分类号： G16B30/00 , C12Q1/6827 , C12Q1/6869 , G06F16/2255 , G06F17/18 , G16B5/00 , G16B10/00 , G16B20/20 , G16B20/40 , G16B40/00 , G16B40/10 , G16B45/00 , G16B50/00
摘要： The technology disclosed generates a reference array of variant data for locations that are shared between read results which are to be compared, and generates hashes over a selected pattern length of positions in the reference array to independently produce non-unique window hashes for base patterns in the read results. It then selects for comparison window hashes that occur less than a ceiling number of times and compares the selected window hashes to identify common window hashes between the read results. It then determines a similarity measure for the read results based on the common window hashes.
-
公开(公告)号：US11755709B2
公开(公告)日：2023-09-12
申请号：US17676797
申请日：2022-02-21
申请人： SHARECARE AI, INC.
发明人： Axel Sly , Srivatsa Akshay Sharma , Brett Robert Redinger , Devin Daniel Reich , Geert Trooskens , Meelis Lootus , Young Jin Lee , Ricardo Lopez Arredondo , Frederick Franklin Kautz, IV , Satish Srinivasan Bhat , Scott Michael Kirk , Walter Adolf De Brouwer , Kartik Thakore
IPC分类号： G06F21/32 , G06F21/45 , G06N20/00 , H04L9/32 , G16H10/60 , G06K7/14 , G06N5/04 , H04L9/08 , G06V40/70 , G06K19/06 , G06F18/214 , G06V10/74 , G06V10/77 , G06V10/80 , G06V10/44 , G06V40/16 , G06V10/774 , H04L9/40 , G06N3/08 , G06N3/04
CPC分类号： G06F21/32 , G06F18/214 , G06F21/45 , G06K7/1417 , G06K19/06037 , G06N5/04 , G06N20/00 , G06V10/451 , G06V10/761 , G06V10/774 , G06V10/7715 , G06V10/803 , G06V40/161 , G06V40/168 , G06V40/70 , G16H10/60 , H04L9/085 , H04L9/0841 , H04L9/0866 , H04L9/0894 , H04L9/3228 , H04L9/3231 , H04L9/3236 , H04L9/3239 , H04L9/3242 , H04L9/3247 , H04L9/3297 , G06N3/04 , G06N3/08 , H04L63/0861
摘要： The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.
-
3.发明授权公开(公告)号：US11177960B2
公开(公告)日：2021-11-16
申请号：US17235889
申请日：2021-04-20
申请人： Sharecare AI, Inc.
发明人： Axel Sly , Srivatsa Akshay Sharma , Brett Robert Redinger , Devin Daniel Reich , Geert Trooskens , Meelis Lootus , Young Jin Lee , Ricardo Lopez Arredondo , Frederick Franklin Kautz, IV , Satish Srinivasan Bhat , Scott Michael Kirk , Walter Adolf De Brouwer , Kartik Thakore
摘要： The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.
-
公开(公告)号：US11551784B2
公开(公告)日：2023-01-10
申请号：US16575278
申请日：2019-09-18
申请人： SHARECARE AI, INC.
IPC分类号： G16B30/00 , G16B40/00 , G16B10/00 , G06F16/22 , G16B20/20 , G16B45/00 , G16B20/40 , G16B5/00 , G16B40/10 , C12Q1/6827 , C12Q1/6869 , G06F17/18 , G16B50/00
摘要： The technology disclosed generates a reference array of variant data for locations that are shared between read results which are to be compared, and generates hashes over a selected pattern length of positions in the reference array to independently produce non-unique window hashes for base patterns in the read results. It then selects for comparison window hashes that occur less than a ceiling number of times and compares the selected window hashes to identify common window hashes between the read results. It then determines a similarity measure for the read results based on the common window hashes.
-
公开(公告)号：US11984198B2
公开(公告)日：2024-05-14
申请号：US18152118
申请日：2023-01-09
申请人： SHARECARE AI, INC.
IPC分类号： G16B30/00 , C12Q1/6827 , C12Q1/6869 , G06F16/22 , G06F17/18 , G16B5/00 , G16B10/00 , G16B20/00 , G16B20/20 , G16B20/40 , G16B40/00 , G16B40/10 , G16B40/30 , G16B45/00 , G16B50/00 , H04L9/32
CPC分类号： G16B30/00 , C12Q1/6827 , C12Q1/6869 , G06F16/2255 , G06F17/18 , G16B5/00 , G16B10/00 , G16B20/20 , G16B20/40 , G16B40/00 , G16B40/10 , G16B45/00 , G16B50/00
摘要： First and second sequenced outputs are accessed. The sequenced outputs contain variants occurring at different carriers and at different carrier positions. Hashes are generated over a selected pattern length of positions for those carrier positions that are shared between the sequenced outputs to produce window hashes for base patterns in first and second sequences. Each sequence is based on the shared carrier positions and the respective sequenced output. The window hashes are non-unique. Window hashes that occur less than a ceiling number times are selected. The selected window hashes are compared between the sequences on a starting position basis such that selected window hashes for base patterns having same start positions in the sequenced outputs are compared. Common window hashes are identified between the sequences based on the comparing. A similarity measure is determined between the sequences based on the common window hashes.
-
公开(公告)号：US11699504B2
公开(公告)日：2023-07-11
申请号：US16575276
申请日：2019-09-18
申请人： SHARECARE AI, INC.
IPC分类号： G06F16/00 , G16B30/00 , G16B40/00 , G16B10/00 , G06F16/22 , G16B20/20 , G16B45/00 , G16B20/40 , G16B5/00 , G16B40/10 , C12Q1/6827 , C12Q1/6869 , G06F17/18 , G16B50/00
CPC分类号： G16B30/00 , C12Q1/6827 , C12Q1/6869 , G06F16/2255 , G06F17/18 , G16B5/00 , G16B10/00 , G16B20/20 , G16B20/40 , G16B40/00 , G16B40/10 , G16B45/00 , G16B50/00
摘要： The technology disclosed generates a reference array of variant data for locations that are shared between read results which are to be compared, and generates hashes over a selected pattern length of positions in the reference array to independently produce non-unique window hashes for base patterns in the read results. It then selects for comparison window hashes that occur less than a ceiling number of times and compares the selected window hashes to identify common window hashes between the read results. It then determines a similarity measure for the read results based on the common window hashes.
-
7.发明授权公开(公告)号：US11321447B2
公开(公告)日：2022-05-03
申请号：US17235876
申请日：2021-04-20
申请人： SHARECARE AI, INC.
发明人： Axel Sly , Srivatsa Akshay Sharma , Brett Robert Redinger , Devin Daniel Reich , Geert Trooskens , Meelis Lootus , Young Jin Lee , Ricardo Lopez Arredondo , Frederick Franklin Kautz, IV , Satish Srinivasan Bhat , Scott Michael Kirk , Walter Adolf De Brouwer , Kartik Thakore
IPC分类号： G06K9/00 , G06F21/45 , G06N20/00 , H04L9/32 , G16H10/60 , G06F21/32 , G06K9/62 , G06K7/14 , G06N5/04 , H04L9/08 , H04L29/06 , G06N3/08 , G06N3/04
摘要： The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.
-
-
-
-
-
-
搜索结果
7 条记录 (耗时 0.015 秒)
