Patents by Inventor Meelis Lootus

Meelis Lootus has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11755709
    Abstract: 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.
    Type: Grant
    Filed: February 21, 2022
    Date of Patent: September 12, 2023
    Assignee: SHARECARE AI, INC.
    Inventors: 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
  • Patent number: 11481688
    Abstract: The technology disclosed relates to systems and methods of cross-platform programming of tiny machine learning (ML) applications. The method includes providing a first declarative instruction that, when processed, interacts with a cross-platform capability of tiny ML hardware. The method includes providing a second declarative instruction that, when processed, invokes at least one procedure block. The method includes providing a third declarative instruction that, when processed, causes output from the tiny ML hardware. The method includes compiling the ML procedure block and the tiny ML model into bytecode. The bytecode interacts, via a virtual machine (VM) layer, with the capability to produce the output responsive to the first, second and third declarative instructions.
    Type: Grant
    Filed: November 10, 2021
    Date of Patent: October 25, 2022
    Assignee: Hammer of the Gods Inc.
    Inventors: Kartik Thakore, Srivatsa Akshay Sharma, Walter Adolf De Brouwer, Geert Trooskens, Meelis Lootus, Sam Leroux, Holly Ly
  • Publication number: 20220337418
    Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic biometric identifiers. The method includes generating a scannable code upon receiving a success nonce from a registration server. The registration server can access a user identifier and a hash of at least a signature using the success nonce. The signature can be generated based at least in part upon a biometric identifier of a user. The method includes recreating the hash of the signature stored by the registration server. The method includes generating the scannable code by encrypting the success nonce and the recreated hash. The biometric identifier of the user is generated by feeding a plurality of non-deterministic biometric inputs to a trained machine learning model producing a plurality of feature vectors. The method includes projecting the plurality of feature vectors onto a surface of a unit hyper-sphere and computing a characteristic identity vector representing the user.
    Type: Application
    Filed: November 15, 2021
    Publication date: October 20, 2022
    Applicant: Sharecare AI, Inc.
    Inventors: 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
  • Patent number: 11455163
    Abstract: The technology disclosed relates to systems and methods for deploying cross-platform applications to tiny ML hardware. The system provides tools to maintain definitions of a first, a second and a third declarative instruction. The system provides tools to maintain compiled bytecode for a procedure block and a tiny ML model that runs on the tiny ML hardware. The compiled bytecode further includes a manifest of one or more capabilities, one or more procedure blocks and at least one output supported by the tiny ML model. The system provides a loader that connects to an instance of the tiny ML hardware. The loader includes logic to verify that the instance of the ML hardware supports the one or more capabilities, the one or more procedure blocks and at least one output specified in the manifest. The loader loads the bytecode and verifies integrity of the load.
    Type: Grant
    Filed: November 10, 2021
    Date of Patent: September 27, 2022
    Assignee: Hammer of the Gods Inc.
    Inventors: Kartik Thakore, Srivatsa Akshay Sharma, Walter Adolf De Brouwer, Geert Trooskens, Meelis Lootus, Sam Leroux, Holly Ly
  • Publication number: 20220269771
    Abstract: 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.
    Type: Application
    Filed: May 2, 2022
    Publication date: August 25, 2022
    Applicant: SHARECARE AI, INC.
    Inventors: 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
  • Publication number: 20220179943
    Abstract: 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.
    Type: Application
    Filed: February 21, 2022
    Publication date: June 9, 2022
    Applicant: SHARECARE AI, INC.
    Inventors: 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
  • Publication number: 20220147341
    Abstract: The technology disclosed relates to systems and methods for deploying cross-platform applications to tiny ML hardware. The system provides tools to maintain definitions of a first, a second and a third declarative instruction. The system provides tools to maintain compiled bytecode for a procedure block and a tiny ML model that runs on the tiny ML hardware. The compiled bytecode further includes a manifest of one or more capabilities, one or more procedure blocks and at least one output supported by the tiny ML model. The system provides a loader that connects to an instance of the tiny ML hardware. The loader includes logic to verify that the instance of the ML hardware supports the one or more capabilities, the one or more procedure blocks and at least one output specified in the manifest. The loader loads the bytecode and verifies integrity of the load.
    Type: Application
    Filed: November 10, 2021
    Publication date: May 12, 2022
    Applicant: Hammer of the Gods Inc., dba HOT-G
    Inventors: Kartik THAKORE, Srivatsa Akshay SHARMA, Walter Adolf DE BROUWER, Geert TROOSKENS, Meelis LOOTUS, Sam LEROUX, Holly LY
  • Publication number: 20220147874
    Abstract: The technology disclosed relates to systems and methods of cross-platform programming of tiny machine learning (ML) applications. The method includes providing a first declarative instruction that, when processed, interacts with a cross-platform capability of tiny ML hardware. The method includes providing a second declarative instruction that, when processed, invokes at least one procedure block. The method includes providing a third declarative instruction that, when processed, causes output from the tiny ML hardware. The method includes compiling the ML procedure block and the tiny ML model into bytecode. The bytecode interacts, via a virtual machine (VM) layer, with the capability to produce the output responsive to the first, second and third declarative instructions.
    Type: Application
    Filed: November 10, 2021
    Publication date: May 12, 2022
    Applicant: Hammer of the Gods Inc., dba HOT-G
    Inventors: Kartik THAKORE, Srivatsa Akshay SHARMA, Walter Adolf DE BROUWER, Geert TROOSKENS, Meelis LOOTUS, Sam LEROUX, Holly LY
  • Patent number: 11321447
    Abstract: 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.
    Type: Grant
    Filed: April 20, 2021
    Date of Patent: May 3, 2022
    Assignee: SHARECARE AI, INC.
    Inventors: 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
  • Patent number: 11256801
    Abstract: 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.
    Type: Grant
    Filed: April 20, 2021
    Date of Patent: February 22, 2022
    Assignee: doc.ai, Inc.
    Inventors: 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
  • Patent number: 11177960
    Abstract: 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.
    Type: Grant
    Filed: April 20, 2021
    Date of Patent: November 16, 2021
    Assignee: Sharecare AI, Inc.
    Inventors: 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
  • Publication number: 20210326433
    Abstract: 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.
    Type: Application
    Filed: April 20, 2021
    Publication date: October 21, 2021
    Applicant: doc.ai, Inc.
    Inventors: 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
  • Publication number: 20210326422
    Abstract: 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.
    Type: Application
    Filed: April 20, 2021
    Publication date: October 21, 2021
    Applicant: doc.ai, Inc.
    Inventors: 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
  • Publication number: 20210328801
    Abstract: 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.
    Type: Application
    Filed: April 20, 2021
    Publication date: October 21, 2021
    Applicant: doc.ai, Inc.
    Inventors: 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
  • Patent number: 8508449
    Abstract: A method and apparatus is provided for reducing color shift in relation to viewing angle in an LCD. The method includes receiving a plurality of pixel data constituting an image, each pixel data including a plurality of sub-pixel color components having respective data values; for each of the pixel data, comparing the sub-pixel color component data values included therein; and based on the comparison, modifying the sub-pixel color component data values included in the pixel data to reduce color shift when displayed on the LCD.
    Type: Grant
    Filed: December 15, 2009
    Date of Patent: August 13, 2013
    Assignee: Sharp Corporation
    Inventors: Benjamin John Broughton, Harry Garth Walton, Paul Antony Gass, Meelis Lootus, Charlotte Wendy Michele Borgers
  • Patent number: 8353617
    Abstract: A backlight is provided for illuminating an at least partially transmissive display. The backlight includes a light source. A light guide receives the light from an edge surface and guides the light by total internal reflection. The light extracted from the lightguide has an angular emission profile such that when incident on a prism film, collimated light is produced.
    Type: Grant
    Filed: September 3, 2009
    Date of Patent: January 15, 2013
    Assignee: Sharp Kabushiki Kaisha
    Inventors: David James Montgomery, Ioannis Papakonstantinou, Jonathan Mather, Meelis Lootus
  • Publication number: 20120169790
    Abstract: A method of operating a display device comprising a display panel comprises receiving main and side image pixel data respectively representing a main image (S1) and a side image (S2). For each of a plurality of pixel groups (S3), where each pixel group comprises at least one pixel of the main image pixel data and at least one spatially corresponding pixel of the side image pixel data, a predetermined mapping is performed (S4) using the pixel data of the pixel group as input. The mapping holds output pixel data for the input pixel data which is known to produce an average on-axis luminance with substantially no dependence on the side image pixel data of the group and an average off-axis luminance with substantially no dependence on the main image pixel data of the group. The signals used to drive the display panel are determined from the output pixel data (S5).
    Type: Application
    Filed: September 16, 2010
    Publication date: July 5, 2012
    Applicant: SHARP KABUSHIKI KAISHA
    Inventors: Benjamin John Broughton, Meelis Lootus, Kenji Maeda, Tatsuo Watanabe, Yoshimitsu Inamori, Takashi Yasumoto
  • Publication number: 20110261093
    Abstract: A method and apparatus is provided for reducing colour shift in relation to viewing angle in an LCD. The method includes receiving a plurality of pixel data constituting an image, each pixel data including a plurality of sub-pixel colour components having respective data values; for each of the pixel data, comparing the sub-pixel colour component data values included therein; and based on the comparison, modifying the sub-pixel colour component data values included in the pixel data with respect to two or more of the plurality of sub-pixel colour components to reduce colour shift when displayed on the LCD.
    Type: Application
    Filed: December 16, 2009
    Publication date: October 27, 2011
    Applicant: SHARP KABUSHIKI KAISHA
    Inventors: Benjamin John Broughton, Harry Garth Walton, Paul Antony Gass, Meelis Lootus, Charlotte Wendy Michele Borgers
  • Publication number: 20110051460
    Abstract: A backlight is provided for illuminating an at least partially transmissive display. The backlight includes a light source. A light guide receives the light from an edge surface and guides the light by total internal reflection. The light extracted from the lightguide has an angular emission profile such that when incident on a prism film, collimated light is produced.
    Type: Application
    Filed: September 3, 2009
    Publication date: March 3, 2011
    Inventors: David James Montgomery, Ioannis Papakonstantinou, Jonathan Mather, Meelis Lootus
  • Publication number: 20100156774
    Abstract: A method and apparatus is provided for reducing colour shift in relation to viewing angle in an LCD. The method includes receiving a plurality of pixel data constituting an image, each pixel data including a plurality of sub-pixel colour components having respective data values; for each of the pixel data, comparing the sub-pixel colour component data values included therein; and based on the comparison, modifying the sub-pixel colour component data values included in the pixel data to reduce colour shift when displayed on the LCD.
    Type: Application
    Filed: December 15, 2009
    Publication date: June 24, 2010
    Inventors: Benjamin BROUGHTON, Harry Garth WALTON, Paul Antony GASS, Meelis LOOTUS, Charlotte Wendy Michele BORGERS