Patents by Inventor Eugene Epshteyn

Eugene Epshteyn 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).

  • Method and system for detecting semantic events
    Patent number: 7324984
    Abstract: A method and system is provided for detecting occurrences of semantic temporal events based on observations extracted from input data and event models. The input data is fed into the system from some data source. Based on specified event to be detected, multiple-layer models corresponding to the event are retrieved. The models are used to determine the types of temporal observations to be extracted from the input data. The extracted temporal observations are then used, in combination with the multiple-layer models of the event, to detect the occurrences of the event.
    Type: Grant
    Filed: November 14, 2003
    Date of Patent: January 29, 2008
    Assignee: Intel Corporation
    Inventors: Vasanth R. Tovinkere, Eugene Epshteyn, Richard Qian
  • Method and system for detecting semantic events
    Patent number: 7177861
    Abstract: A method and system is provided for detecting occurrences of semantic temporal events based on observations extracted from input data and event models. The input data is fed into the system from some data source. Based on specified event to be detected, multiple-layer models corresponding to the event are retrieved. The models are used to determine the types of temporal observations to be extracted from the input data. The extracted temporal observations are then used, in combination with the multiple-layer models of the event, to detect the occurrences of the event.
    Type: Grant
    Filed: November 14, 2003
    Date of Patent: February 13, 2007
    Assignee: Intel Corporation
    Inventors: Vasanth R. Tovinkere, Eugene Epshteyn, Richard J. Qian
  • Collecting event data and describing events
    Patent number: 6810397
    Abstract: A method of and a system for collecting, data about an event and describing the event, which for example, might be an entertainment event, a sports event, or a political event. Identifying characteristics of the event, characteristics of types of data about the event, and a record of the event are acquired by a description generator. A person wishing to view a particular type of data utilizes an interactive receiver to select the type of data, and characteristics of the selected type of data within the record of the event are located so as to locate data of the selected type. The located data are then collected and displayed. By storing characteristics of different types of segments that occur in the event, a viewer can select segments of a particular type for viewing. Characteristics of the selected type of segments are then located within the record of the event so as to locate segments of the selected type, and the located segments are then displayed.
    Type: Grant
    Filed: June 29, 2000
    Date of Patent: October 26, 2004
    Assignee: Intel Corporation
    Inventors: Richard J. Qian, Eugene Epshteyn
  • Method and system for detecting semantic events
    Publication number: 20040153288
    Abstract: A method and system is provided for detecting occurrences of semantic temporal events based on observations extracted from input data and event models. The input data is fed into the system from some data source. Based on specified event to be detected, multiple-layer models corresponding to the event are retrieved. The models are used to determine the types of temporal observations to be extracted from the input data. The extracted temporal observations are then used, in combination with the multiple-layer models of the event, to detect the occurrences of the event.
    Type: Application
    Filed: November 14, 2003
    Publication date: August 5, 2004
    Applicant: INTEL CORPORATION
    Inventors: Vasanth R. Tovinkere, Eugene Epshteyn, Richard Qian
  • Method and system for detecting semantic events
    Publication number: 20040102921
    Abstract: A method and system is provided for detecting occurrences of semantic temporal events based on observations extracted from input data and event models. The input data is fed into the system from some data source. Based on specified event to be detected, multiple-layer models corresponding to the event are retrieved. The models are used to determine the types of temporal observations to be extracted from the input data. The extracted temporal observations are then used, in combination with the multiple-layer models of the event, to detect the occurrences of the event.
    Type: Application
    Filed: November 14, 2003
    Publication date: May 27, 2004
    Applicant: INTEL CORPORATION
    Inventors: Vasanth R. Tovinkere, Eugene Epshteyn, Richard J. Quian
  • Method and system for detecting semantic events
    Patent number: 6678635
    Abstract: A method and system is provided for detecting occurrences of semantic temporal events based on observations extracted from input data and event models. The input data is fed into the system from some data source. Based on specified event to be detected, multiple-layer models corresponding to the event are retrieved. The models are used to determine the types of temporal observations to be extracted from the input data. The extracted temporal observations are then used, in combination with the multiple-layer models of the event, to detect the occurrences of the event.
    Type: Grant
    Filed: January 23, 2001
    Date of Patent: January 13, 2004
    Assignee: Intel Corporation
    Inventors: Vasanth R. Tovinkere, Eugene Epshteyn, Richard J. Qian
  • Method and system for detecting semantic events
    Publication number: 20020099518
    Abstract: A method and system is provided for detecting occurrences of semantic temporal events based on observations extracted from input data and event models. The input data is fed into the system from some data source. Based on specified event to be detected, multiple-layer models corresponding to the event are retrieved. The models are used to determine the types of temporal observations to be extracted from the input data. The extracted temporal observations are then used, in combination with the multiple-layer models of the event, to detect the occurrences of the event.
    Type: Application
    Filed: January 23, 2001
    Publication date: July 25, 2002
    Inventors: Vasanth R. Tovinkere, Eugene Epshteyn, Richard J. Qian