Patents by Inventor Adam Waldemar Golinski

Adam Waldemar Golinski 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: 11405626
    Abstract: Techniques are described herein for coding video content using recurrent-based machine learning tools. A device can include a neural network system including encoder and decoder portions. The encoder portion can generate output data for the current time step of operation of the neural network system based on an input video frame for a current time step of operation of the neural network system, reconstructed motion estimation data from a previous time step of operation, reconstructed residual data from the previous time step of operation, and recurrent state data from at least one recurrent layer of a decoder portion of the neural network system from the previous time step of operation. A decoder portion of the neural network system can generate, based on the output data and recurrent state data from the previous time step of operation, a reconstructed video frame for the current time step of operation.
    Type: Grant
    Filed: November 6, 2020
    Date of Patent: August 2, 2022
    Assignee: QUALCOMM Incorporated
    Inventors: Adam Waldemar Golinski, Yang Yang, Reza Pourreza, Guillaume Konrad Sautiere, Ties Jehan Van Rozendaal, Taco Sebastiaan Cohen
  • Publication number: 20210281867
    Abstract: Techniques are described herein for coding video content using recurrent-based machine learning tools. A device can include a neural network system including encoder and decoder portions. The encoder portion can generate output data for the current time step of operation of the neural network system based on an input video frame for a current time step of operation of the neural network system, reconstructed motion estimation data from a previous time step of operation, reconstructed residual data from the previous time step of operation, and recurrent state data from at least one recurrent layer of a decoder portion of the neural network system from the previous time step of operation. A decoder portion of the neural network system can generate, based on the output data and recurrent state data from the previous time step of operation, a reconstructed video frame for the current time step of operation.
    Type: Application
    Filed: November 6, 2020
    Publication date: September 9, 2021
    Inventors: Adam Waldemar GOLINSKI, Yang YANG, Reza POURREZA, Guillaume Konrad SAUTIERE, Ties Jehan VAN ROZENDAAL, Taco Sebastiaan COHEN
  • Publication number: 20170316683
    Abstract: In some embodiments, a sensor device can include a base module including a battery and including a transceiver configured to communicate with a computing device. The sensor device may further include one or more sensor modules configured to releasably couple to the base module.
    Type: Application
    Filed: February 13, 2017
    Publication date: November 2, 2017
    Applicant: IDEATION SYSTEMS LLC
    Inventors: Lukasz Leszek Pietrasik, Adam Waldemar Golinski, Filip Mateusz Kaklin, Ahmad Hani Zaatari, Daniel James Yee, Mohsen Nakhaeinejad
  • Publication number: 20160305797
    Abstract: In some embodiments, a sensor apparatus may include a base module having a first interface configured to receive a signal from at least one sensor module. The base module may also include a controller coupled to the first interface and configured to process the signal to produce data, and may include a second interface configured to communicate the data to a computing device.
    Type: Application
    Filed: February 13, 2016
    Publication date: October 20, 2016
    Applicant: Ideation Systems LLC
    Inventors: Lukasz Leszek Pietrasik, Adam Waldemar Golinski, Filip Mateusz Kaklin, Ahmad Hani Zaatari, Daniel James Yee, Mohsen Nakhaeinejad