Patents by Inventor John J. Guo
John J. Guo 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: 11350105Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: GrantFiled: January 8, 2021Date of Patent: May 31, 2022Assignee: Euclid Discoveries, LLCInventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Patent number: 11228766Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: GrantFiled: January 7, 2021Date of Patent: January 18, 2022Assignee: EUCLID DISCOVERIES, LLCInventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Patent number: 11159801Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: GrantFiled: July 10, 2020Date of Patent: October 26, 2021Assignee: EUCLID DISCOVERIES, LLCInventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Publication number: 20210203950Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: ApplicationFiled: January 7, 2021Publication date: July 1, 2021Inventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Publication number: 20210203951Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: ApplicationFiled: January 8, 2021Publication date: July 1, 2021Inventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Publication number: 20200413067Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: ApplicationFiled: July 10, 2020Publication date: December 31, 2020Inventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Patent number: 10757419Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: GrantFiled: May 23, 2019Date of Patent: August 25, 2020Assignee: Euclid Discoveries, LLCInventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Publication number: 20190289296Abstract: Videos may be characterized by objective metrics that quantify video quality. Embodiments are directed to target bitrate prediction methods in which one or more objective metrics may serve as inputs into a model that predicts a mean opinion score (MOS), a measure of perceptual quality, as a function of metric values. The model may be derived by generating training data through conducting subjective tests on a set of video encodings, obtaining MOS data from the subjective tests, and correlating the MOS data with metric measurements on the training data. The MOS predictions may be extended to predict the target (encoding) bitrate that achieves a desired MOS value. The target bitrate prediction methods may be applied to segments of a video. The methods may be made computationally faster by applying temporal subsampling.Type: ApplicationFiled: May 23, 2019Publication date: September 19, 2019Inventors: Dane P. Kottke, Katherine H. Cornog, John J. Guo, Myo Tun, Jeyun Lee, Nigel Lee
-
Patent number: 9621917Abstract: Continuous block tracking (CBT) tracks macroblock locations over reference frames to produce better inter-predictions than conventional block-based motion estimation/compression. CBT includes frame-to-frame tracking, estimating motion from a frame to a previous frame, and continuous tracking, related frame-to-frame motion vectors to block tracks. Frame-to-frame tracking may include block based or hierarchical motion estimations. CBT combined with enhanced predictive zonal search may create unified motion estimation. Accumulated CBT results may form trajectories for trajectory-based CBT predictions. Metrics measuring continuous track and motion vector quality can assess relative priority of CBT prediction against non-tracker-based predictions and to modify encoding choices. Continuous tracks can be analyzed for goodness-of-fit to translational motion models, with outliers removed from encoding. Translational motion models can be extended to entire frames in adaptive picture type selection.Type: GrantFiled: November 4, 2014Date of Patent: April 11, 2017Assignee: EUCLID DISCOVERIES, LLCInventors: Dane P. Kottke, John J. Guo, Jeyun Lee, Sangseok Park, Christopher Weed, Justin Kwan, Nigel Lee
-
Publication number: 20150256850Abstract: Continuous block tracking (CBT) tracks macroblock locations over reference frames to produce better inter-predictions than conventional block-based motion estimation/compression. CBT includes frame-to-frame tracking, estimating motion from a frame to a previous frame, and continuous tracking, related frame-to-frame motion vectors to block tracks. Frame-to-frame tracking may include block based or hierarchical motion estimations. CBT combined with enhanced predictive zonal search may create unified motion estimation. Accumulated CBT results may form trajectories for trajectory-based CBT predictions. Metrics measuring continuous track and motion vectors quality can assess relative priority of CBT predictions against non-tracker-based predictions and to modify encoding choices. Continuous tracks can be analyzed for goodness-of-fit to translational motion models, with outliers removed from encoding. Translational motion models can be extended to entire frames in adaptive picture type selection.Type: ApplicationFiled: November 4, 2014Publication date: September 10, 2015Inventors: Dane P. Kottke, John J. Guo, Jeyun Lee, Sangseok Park, Christopher Weed, Justin Kwan, Nigel Lee
-
Patent number: 8134449Abstract: A method for training a computing system using keyboard biometric information. The method includes depressing two or more keys on a keyboard input device for a first sequence of keys. The method then determines a key press time for each of the two or more keys to provide a key press time characteristic in the first sequence of keys. The method also determines a flight time between a first key and a second key to provide a flight time characteristic in the first sequence of keys, the first key being within the two or more keys. The method includes storing the key press time characteristic and the flight time characteristic for the first sequence of keys, and displaying indications associated with the first sequence of keys on a display device provided on a portion of the keyboard input device.Type: GrantFiled: October 22, 2008Date of Patent: March 13, 2012Assignee: Minebea Co., LtdInventors: Mario T. Wu, Jr., John J. Guo, Larry Rice
-
Publication number: 20090134972Abstract: A method for training a computing system using keyboard biometric information. The method includes depressing two or more keys on a keyboard input device for a first sequence of keys. The method then determines a key press time for each of the two or more keys to provide a key press time characteristic in the first sequence of keys. The method also determines a flight time between a first key and a second key to provide a flight time characteristic in the first sequence of keys, the first key being within the two or more keys. The method includes storing the key press time characteristic and the flight time characteristic for the first sequence of keys, and displaying indications associated with the first sequence of keys on a display device provided on a portion of the keyboard input device.Type: ApplicationFiled: October 22, 2008Publication date: May 28, 2009Applicant: Minebea Co., Ltd.Inventors: MARIO T. WU, JR., John J. Guo, Larry Rice