Patents by Inventor Jacob Borgman
Jacob Borgman 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).
-
Publication number: 20230259767Abstract: The disclosure relates to a system for evaluating movement of a body of a user. The system may include a video display, one or more digital cameras, and a processor. The processor may control the one or more cameras to generate images of at least the part of the body over a period of time. The processor may estimate a position of a plurality of joints of the body. The processor may receive a selection of a tracked pose, and determine, from the plurality of joints, a set of joints associated with the tracked pose. The processor may generate at least one joint vector connecting joints in the set of joints, and assign, based on changes in the joint vector over the period of time, a form score to a performance of the tracked pose. The processor may then generate a user interface that depicts the form score.Type: ApplicationFiled: April 24, 2023Publication date: August 17, 2023Inventors: Hemant Virkar, Leah R. Kaplan, Stephen Furlani, Jacob Borgman, Anil Bhave, Mihir Thakkar, Sunkist Mehta
-
Patent number: 11633659Abstract: The disclosure relates to a system for evaluating movement of a body of a user. The system may include a video display, one or more digital cameras, and a processor. The processor may control the one or more cameras to generate images of at least the part of the body over a period of time. The processor may estimate a position of a plurality of joints of the body. The processor may receive a selection of a tracked pose, and determine, from the plurality of joints, a set of joints associated with the tracked pose. The processor may generate at least one joint vector connecting joints in the set of joints, and assign, based on changes in the joint vector over the period of time, a form score to a performance of the tracked pose. The processor may then generate a user interface that depicts the form score.Type: GrantFiled: June 29, 2021Date of Patent: April 25, 2023Assignee: MirrorAR LLCInventors: Hemant Virkar, Leah R. Kaplan, Stephen Furlani, Jacob Borgman, Anil Bhave
-
Publication number: 20210322856Abstract: The disclosure relates to a system for evaluating movement of a body of a user. The system may include a video display, one or more digital cameras, and a processor. The processor may control the one or more cameras to generate images of at least the part of the body over a period of time. The processor may estimate a position of a plurality of joints of the body. The processor may receive a selection of a tracked pose, and determine, from the plurality of joints, a set of joints associated with the tracked pose. The processor may generate at least one joint vector connecting joints in the set of joints, and assign, based on changes in the joint vector over the period of time, a form score to a performance of the tracked pose. The processor may then generate a user interface that depicts the form score.Type: ApplicationFiled: June 29, 2021Publication date: October 21, 2021Inventors: Hemant Virkar, Leah R. Kaplan, Stephen Furlani, Jacob Borgman, Anil Bhave
-
Patent number: 11069144Abstract: The disclosure relates to a system for providing guidance for positioning a body. The system may include a video display, one or more digital cameras configured to generate a depth video stream and a visual video stream, and a computing device including, a memory, and a processor. The processor may control the one or more digital cameras to generate the depth video stream including a depth image of the body and the visual video stream including a color image of the body. The processor identifies at least a part of the body within the images using a first trained learning machine to segment the images and isolate the body. The processor may crop both the visual image and the depth image based on the identified body. The processor may estimate a position of a plurality of joints of the body by applying a second trained learning machine to the identified and isolated part of the body. The processor may generate a current pose estimate by connecting estimated positions of the plurality of joints.Type: GrantFiled: September 13, 2019Date of Patent: July 20, 2021Assignee: MirrorAR LLCInventors: Hemant Virkar, Leah Kaplan, Stephen Furlani, Jacob Borgman
-
Patent number: 10402748Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.Type: GrantFiled: June 5, 2015Date of Patent: September 3, 2019Assignee: HEMANT V. VIRKARInventors: Hemant Virkar, Karen Stark, Jacob Borgman
-
Publication number: 20150286955Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.Type: ApplicationFiled: June 5, 2015Publication date: October 8, 2015Inventors: Hemant VIRKAR, Karen STARK, Jacob BORGMAN
-
Patent number: 9082083Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.Type: GrantFiled: January 22, 2013Date of Patent: July 14, 2015Assignee: DIGITAL INFUZION, INC.Inventors: Hemant Virkar, Karen Stark, Jacob Borgman
-
Patent number: 8812274Abstract: Methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces, and to machines and systems relating thereto. More specifically, exemplary aspects of the invention relate to methods and systems for generating supervised hypersurfaces based on user domain expertise, machine learning techniques, or other supervised learning techniques. These supervised hypersurfaces may optionally be combined with unsupervised hypersurfaces derived from unsupervised learning techniques. Lower-dimensional subspaces may be determined by the methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces. Data may then be projected onto the lower-dimensional subspaces for use, e.g., in further data discovery, visualization for display, or database access.Type: GrantFiled: April 26, 2010Date of Patent: August 19, 2014Inventors: Hemant Virkar, Karen Stark, Jacob Borgman
-
Publication number: 20130238533Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.Type: ApplicationFiled: January 22, 2013Publication date: September 12, 2013Applicant: Digital Infuzion, Inc.Inventors: Hemant VIRKAR, Karen Stark, Jacob Borgman
-
Patent number: 8386401Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.Type: GrantFiled: September 10, 2009Date of Patent: February 26, 2013Assignee: Digital Infuzion, Inc.Inventors: Hemant Virkar, Karen Stark, Jacob Borgman
-
Publication number: 20100274539Abstract: Methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces, and to machines and systems relating thereto. More specifically, exemplary aspects of the invention relate to methods and systems for generating supervised hypersurfaces based on user domain expertise, machine learning techniques, or other supervised learning techniques. These supervised hypersurfaces may optionally be combined with unsupervised hypersurfaces derived from unsupervised learning techniques. Lower-dimensional subspaces may be determined by the methods and systems for creating ensembles of hypersurfaces in high-dimensional feature spaces. Data may then be projected onto the lower-dimensional subspaces for use, e.g., in further data discovery, visualization for display, or database access.Type: ApplicationFiled: April 26, 2010Publication date: October 28, 2010Inventors: Hemant VIRKAR, Karen STARK, Jacob BORGMAN
-
Publication number: 20100063948Abstract: Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples.Type: ApplicationFiled: September 10, 2009Publication date: March 11, 2010Applicant: DIGITAL INFUZION, INC.Inventors: Hemant VIRKAR, Karen Stark, Jacob Borgman