-
1.发明公开公开(公告)号：US20240355084A1
公开(公告)日：2024-10-24
申请号：US18751930
申请日：2024-06-24
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Joan Devassy , Nadha Nafeeza Gafoor , Aahel Iyer , Swarn Avinash Kumar , Angela Lam , Kia Nishimine , Wiebke Poerschke , John Michael Sparks , Hristo Stefanov Stefanov , Wei You
IPC: G06V10/24 , G06F18/214 , G06N20/00 , G06Q20/34 , G06V10/25
CPC classification number: G06V10/242 , G06F18/2148 , G06N20/00 , G06Q20/353 , G06Q20/3552 , G06V10/25
Abstract: This disclosure describes a card-scan system that can update a card-scan machine learning model to improve card-character predictions for character-bearing cards by using an active-learning technique that learns from card-scan representations indicating corrections by users to predicted card characters. In particular, the disclosed systems can use a client device to capture and analyze a set of card images of a character-bearing card to predict card characters using a card-scan machine learning model. The disclosed systems can further receive card-scan gradients representing one or more corrections to incorrectly predicted card characters. Based on the card-scan gradients, the disclosed systems can generate active-learning metrics and retrain or update the card-scan machine learning model based on such active-learning metrics. The disclosed systems can improve the accuracy with which card-character-detection systems predict card characters while preserving data security and verifying the presence of a physical character-bearing card.
-
2.发明申请公开(公告)号：US20220108121A1
公开(公告)日：2022-04-07
申请号：US17063033
申请日：2020-10-05
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Joan Devassy , Nadha Nafeeza Gafoor , Aahel Iyer , Swarn Avinash Kumar , Angela Lam , Kia Nishimine , Wiebke Poerschke , John Michael Sparks , Hristo Stefanov Stefanov , Wei You
Abstract: This disclosure describes a card-scan system that can update a card-scan machine learning model to improve card-character predictions for payment cards, driver licenses, or other character-bearing cards by using an active-learning technique that learns from card-scan representations indicating corrections by users to predicted card characters. In particular, the disclosed systems can use a client device to capture and analyze a set of card images of a character-bearing card to predict card characters using a card-scan machine learning model. The disclosed systems can further receive card-scan gradients representing one or more corrections to incorrectly predicted card characters. Based on the card-scan gradients, the disclosed systems can generate active-learning metrics and retrain or update the card-scan machine learning model based on such active-learning metrics. By utilizing and updating the card-scan machine learning model, the disclosed systems can improve the accuracy with which card-character-detection systems predict card characters while preserving data security and verifying the presence of a physical character-bearing card.
-
公开(公告)号：US12018958B2
公开(公告)日：2024-06-25
申请号：US16917672
申请日：2020-06-30
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Kris Richard Efland , Nadha Nafeeza Gafoor , Nastaran Ghadar , Meng Tao
CPC classification number: G01C21/3602 , G01C21/3492 , G01C21/3833 , G01C21/3896 , G06F18/214 , G06V20/582 , G06V20/584 , G06V20/588
Abstract: Examples disclosed herein may involve a computing system configured to (i) identify a stationary element in a real-world environment for which to infer information, (ii) detect a semantic relationship between the stationary element and one or more other stationary elements in the real-world environment, (iii) based on the detected semantic relationship, infer information about the stationary element, (iv) include the inferred information about the stationary element within a set of information that describes the stationary element.
-
Patent Agency Ranking
公开(公告)号：US12158351B2
公开(公告)日：2024-12-03
申请号：US16917672
申请日：2020-06-30
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Kris Richard Efland , Nadha Nafeeza Gafoor , Nastaran Ghadar , Meng Tao
Abstract: Examples disclosed herein may involve a computing system configured to (i) identify a stationary element in a real-world environment for which to infer information, (ii) detect a semantic relationship between the stationary element and one or more other stationary elements in the real-world environment, (iii) based on the detected semantic relationship, infer information about the stationary element, (iv) include the inferred information about the stationary element within a set of information that describes the stationary element.
-
公开(公告)号：US20210404841A1
公开(公告)日：2021-12-30
申请号：US16917672
申请日：2020-06-30
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Kris Richard Efland , Nadha Nafeeza Gafoor , Nastaran Ghadar , Meng Tao
Abstract: Examples disclosed herein may involve a computing system configured to (i) identify a stationary element in a real-world environment for which to infer information, (ii) detect a semantic relationship between the stationary element and one or more other stationary elements in the real-world environment, (iii) based on the detected semantic relationship, infer information about the stationary element, (iv) include the inferred information about the stationary element within a set of information that describes the stationary element.
-
6.发明授权公开(公告)号：US12026926B2
公开(公告)日：2024-07-02
申请号：US17063033
申请日：2020-10-05
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Joan Devassy , Nadha Nafeeza Gafoor , Aahel Iyer , Swarn Avinash Kumar , Angela Lam , Kia Nishimine , Wiebke Poerschke , John Michael Sparks , Hristo Stefanov Stefanov , Wei You
IPC: G06N20/00 , G06F18/214 , G06Q20/34 , G06V10/24 , G06V10/25
CPC classification number: G06V10/242 , G06F18/2148 , G06N20/00 , G06Q20/353 , G06Q20/3552 , G06V10/25
Abstract: This disclosure describes a card-scan system that can update a card-scan machine learning model to improve card-character predictions for payment cards, driver licenses, or other character-bearing cards by using an active-learning technique that learns from card-scan representations indicating corrections by users to predicted card characters. In particular, the disclosed systems can use a client device to capture and analyze a set of card images of a character-bearing card to predict card characters using a card-scan machine learning model. The disclosed systems can further receive card-scan gradients representing one or more corrections to incorrectly predicted card characters. Based on the card-scan gradients, the disclosed systems can generate active-learning metrics and retrain or update the card-scan machine learning model based on such active-learning metrics. By utilizing and updating the card-scan machine learning model, the disclosed systems can improve the accuracy with which card-character-detection systems predict card characters while preserving data security and verifying the presence of a physical character-bearing card.
-
公开(公告)号：US11682124B2
公开(公告)日：2023-06-20
申请号：US16916029
申请日：2020-06-29
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Kris Richard Efland , Nadha Nafeeza Gafoor , Nastaran Ghadar , Meng Tao
CPC classification number: G06T7/337 , G01C21/3833 , G06T3/0075 , G06T3/20 , G06T3/60 , G06T7/60 , G06T7/74 , G06T11/60
Abstract: Examples disclosed herein may involve a computing system that is operable to (i) identify a source map and a target map for transferring map data, where the source map and the target map have different respective coordinate frames and respective coverage areas that at least partially overlap, (ii) select a real-world element for which to transfer previously-created map data from the source map to the target map, (iii) select a source image associated with the source map in which the selected real-world element appears and has been labeled, (iv) select a target image associated with the target map in which the selected real-world element appears, (v) derive a geometric relationship between the source image and the target image, and (vi) use the derived geometric relationship between the source image and the target image to determine a position of the real-world element within the respective coordinate frame of the target map.
-
公开(公告)号：US20210407114A1
公开(公告)日：2021-12-30
申请号：US16916029
申请日：2020-06-29
Applicant: Lyft, Inc.
Inventor： Ritwik Subir Das , Kris Richard Efland , Nadha Nafeeza Gafoor , Nastaran Ghadar , Meng Tao
Abstract: Examples disclosed herein may involve a computing system that is operable to (i) identify a source map and a target map for transferring map data, where the source map and the target map have different respective coordinate frames and respective coverage areas that at least partially overlap, (ii) select a real-world element for which to transfer previously-created map data from the source map to the target map, (iii) select a source image associated with the source map in which the selected real-world element appears and has been labeled, (iv) select a target image associated with the target map in which the selected real-world element appears, (v) derive a geometric relationship between the source image and the target image, and (vi) use the derived geometric relationship between the source image and the target image to determine a position of the real-world element within the respective coordinate frame of the target map.
-
-
-
-
-
-
-
Search Results
8
records
(took 0.012 seconds)
