Patents by Inventor Karel Jan Willem MOENS

Karel Jan Willem MOENS 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 training generative adversarial networks with heterogeneous data
    Patent number: 11823348
    Abstract: A method and system for training a neural network to perform processing of digital data. The input data can be heterogeneous and the method or system obtains a multiple of loss signals. The input data can be selected so that the loss signals are balanced and can fulfill several conditions of the output data. When running a trained neural network on digital frame images, intermediate results of processing a frame can be used on later frames and in this way processing delay can be decreased.
    Type: Grant
    Filed: May 17, 2019
    Date of Patent: November 21, 2023
    Assignee: BARCO N.V.
    Inventor: Karel Jan Willem Moens
  • METHOD AND SYSTEM FOR TRAINING GENERATIVE ADVERSARIAL NETWORKS WITH HETEROGENEOUS DATA
    Publication number: 20220215504
    Abstract: A method and system for training a neural network to perform processing of digital data. The input data can be heterogeneous and the method or system obtains a multiple of loss signals. The input data can be selected so that the loss signals are balanced and can fulfill several conditions of the output data. When running a trained neural network on digital frame images, intermediate results of processing a frame can be used on later frames and in this way processing delay can be decreased.
    Type: Application
    Filed: May 17, 2019
    Publication date: July 7, 2022
    Applicant: BARCO N.V.
    Inventor: Karel Jan Willem MOENS