Patents by Inventor Ferenc Huszar

Ferenc Huszar 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: 20200280730
    Abstract: Methods and systems for optimising the quality of visual data. Specifically, methods and systems for preserving visual information during compression and decompression. An example method for optimising visual data includes using a pre-processing neural network to optimise visual data prior to encoding the visual data in visual data processing; and using a post-processing neural network to enhance visual data following decoding visual data in visual data processing.
    Type: Application
    Filed: May 13, 2020
    Publication date: September 3, 2020
    Inventors: Zehan Wang, Robert David Bishop, Ferenc Huszar, Lucas Theis
  • Patent number: 10701394
    Abstract: A method includes selecting a plurality of low-resolution frames associated with a video, performing a first motion estimation between a first frame and a second frame, performing a second motion estimation between a third frame and the second frame, generating a high-resolution frame representing the second frame based on the first motion estimation, the second motion estimation and the second frame using a sub-pixel convolutional neural network.
    Type: Grant
    Filed: November 10, 2017
    Date of Patent: June 30, 2020
    Assignee: Twitter, Inc.
    Inventors: Jose Caballero, Christian Ledig, Andrew Aitken, Alfredo Alejandro Acosta Diaz, Lucas Theis, Ferenc Huszar, Johannes Totz, Zehan Wang, Wenzhe Shi
  • Patent number: 10692185
    Abstract: A method for training an algorithm to process at least a section of received visual data using a training dataset and reference dataset. The method comprises an iterative method with iterations comprising: generating a set of training data using the algorithm; comparing one or more characteristics of the training data to one or more characteristics of at least a section of the reference dataset; and modifying one or more parameters of the algorithm to optimise processed visual data based on the comparison between the characteristic of the training data and the characteristic of the reference dataset. The algorithm may output the processed visual data with the same content as the at least a section of received visual data. Some aspects and/or implementations provide for improved super-resolution of lower quality images to produce super-resolution images with improved characteristics (e.g. less blur, less undesired smoothing) compared to other super-resolution techniques.
    Type: Grant
    Filed: December 28, 2017
    Date of Patent: June 23, 2020
    Assignee: Magic Pony Technology Limited
    Inventors: Zehan Wang, Wenzhe Shi, Ferenc Huszar, Robert David Bishop
  • Patent number: 10685264
    Abstract: The present disclosure relates to a method for processing input visual data using a generated algorithm based upon input visual data and the output of a calculated energy function. According to a first aspect of the disclosure, there is provided a method for enhancing input visual data using an algorithm, the method comprising evaluating gradients of the output of an energy function with respect to the input visual data; using the gradient output to enhance the input visual data; and outputting the enhanced visual data.
    Type: Grant
    Filed: December 27, 2017
    Date of Patent: June 16, 2020
    Assignee: Magic Pony Technology Limited
    Inventors: Ferenc Huszar, Robert David Bishop, Zehan Wang
  • Patent number: 10681361
    Abstract: Methods and systems for optimising the quality of visual data. Specifically, methods and systems for preserving visual information during compression and decompression. An example method for optimising visual data includes using a pre-processing hierarchical algorithm to optimise visual data prior to encoding the visual data in visual data processing; and using a post-processing hierarchical algorithm to enhance visual data following decoding visual data in visual data processing.
    Type: Grant
    Filed: December 27, 2017
    Date of Patent: June 9, 2020
    Assignee: Magic Pony Technology Limited
    Inventors: Zehan Wang, Robert David Bishop, Ferenc Huszar, Lucas Theis
  • Patent number: 10666962
    Abstract: Disclosed is method for training a plurality of visual processing algorithms for processing visual data. The method includes using a pre-processing hierarchical algorithm to process the visual data prior to encoding the visual data in visual data processing, and using a post-processing hierarchical algorithm to further process the visual data following decoding visual data in visual data processing. The encoding and decoding are performed with respect to a predetermined visual data codec and may be content specific.
    Type: Grant
    Filed: September 18, 2017
    Date of Patent: May 26, 2020
    Assignee: Magic Pony Technology Limited
    Inventors: Zehan Wang, Robert David Bishop, Ferenc Huszar, Lucas Theis
  • Patent number: 10623775
    Abstract: A system (e.g., an auto-encoder system) includes an encoder, a decoder and a learning module. The encoder generates compressed video data using a lossy compression algorithm, the lossy compression algorithm being implemented using a trained neural network with at least one convolution, generate at least one first parameter based on the compressed video data, and communicate the compressed video data and the model to at least one device configured to decode the compressed video data using an inverse algorithm based on the lossy compression algorithm. The decoder generates decoded video data based on the compressed video data using the inverse algorithm and the model, and generate at least one second parameter based on the decoded video data. The learning module trains the model using the at least one first parameter and the at least one second parameter.
    Type: Grant
    Filed: November 6, 2017
    Date of Patent: April 14, 2020
    Assignee: Twitter, Inc.
    Inventors: Lucas Theis, Ferenc Huszar, Zehan Wang, Wenzhe Shi
  • Publication number: 20180240031
    Abstract: Systems and methods provide a deep neural network trained via active learning. An example method includes generating, from a set of labeled objects, a plurality of differing training sets, assigning each of the plurality of training sets to a respective deep neural network in a committee of networks, and initializing each of the deep neural networks in the committee by training the deep neural network using the respective assigned training set. The method further includes iteratively training the deep neural networks in the committee until convergence and using one of the deep neural networks to make predictions for unlabeled objects. The training may include identifying unlabeled objects with highest diversity in predictions from the plurality of deep neural networks, obtaining a respective label for each identified unlabeled object, and retraining the deep neural networks with the respective labels for the objects.
    Type: Application
    Filed: January 22, 2018
    Publication date: August 23, 2018
    Inventors: Ferenc Huszar, Pietro Berkes, Zehan Wang
  • Publication number: 20180240017
    Abstract: Systems and methods provide a learned difference metric that operates in a wide artifact space. An example method includes initializing a committee of deep neural networks with labeled distortion pairs, iteratively actively learning a difference metric using the committee and psychophysics tasks for informative distortion pairs, and using the difference metric as an objective function in a machine-learned digital file processing task. Iteratively actively learning the difference metric can include providing an unlabeled distortion pair as input to each of the deep neural networks in the committee, a distortion pair being a base image and a distorted image resulting from application of an artifact applied to the base image, obtaining a plurality of difference metric scores for the unlabeled distortion pair from the deep neural networks, and identifying the unlabeled distortion pair as an informative distortion pair when the difference metric scores satisfy a diversity metric.
    Type: Application
    Filed: December 27, 2017
    Publication date: August 23, 2018
    Inventors: Ferenc Huszar, Lucas Theis, Pietro Berkes
  • Publication number: 20180139458
    Abstract: Methods and systems for optimising the quality of visual data. Specifically, methods and systems for preserving visual information during compression and decompression. An example method for optimising visual data includes using a pre-processing hierarchical algorithm to optimise visual data prior to encoding the visual data in visual data processing; and using a post-processing hierarchical algorithm to enhance visual data following decoding visual data in visual data processing.
    Type: Application
    Filed: December 27, 2017
    Publication date: May 17, 2018
    Inventors: Zehan Wang, Robert David Bishop, Ferenc Huszar, Lucas Theis
  • Publication number: 20180131953
    Abstract: Disclosed is method for training a plurality of visual processing algorithms for processing visual data. The method includes using a pre-processing hierarchical algorithm to process the visual data prior to encoding the visual data in visual data processing, and using a post-processing hierarchical algorithm to further process the visual data following decoding visual data in visual data processing. The encoding and decoding are performed with respect to a predetermined visual data codec and may be content specific.
    Type: Application
    Filed: September 18, 2017
    Publication date: May 10, 2018
    Inventors: Zehan Wang, Robert David Bishop, Ferenc Huszar, Lucas Theis
  • Publication number: 20180121769
    Abstract: The present disclosure relates to a method for processing input visual data using a generated algorithm based upon input visual data and the output of a calculated energy function. According to a first aspect of the disclosure, there is provided a method for enhancing input visual data using an algorithm, the method comprising evaluating gradients of the output of an energy function with respect to the input visual data; using the gradient output to enhance the input visual data; and outputting the enhanced visual data.
    Type: Application
    Filed: December 27, 2017
    Publication date: May 3, 2018
    Inventors: Ferenc Huszar, Robert David Bishop, Zehan Wang
  • Publication number: 20180122048
    Abstract: A method for training an algorithm to process at least a section of received visual data using a training dataset and reference dataset. The method comprises an iterative method with iterations comprising: generating a set of training data using the algorithm; comparing one or more characteristics of the training data to one or more characteristics of at least a section of the reference dataset; and modifying one or more parameters of the algorithm to optimise processed visual data based on the comparison between the characteristic of the training data and the characteristic of the reference dataset. The algorithm may output the processed visual data with the same content as the at least a section of received visual data. Some aspects and/or implementations provide for improved super-resolution of lower quality images to produce super-resolution images with improved characteristics (e.g. less blur, less undesired smoothing) compared to other super-resolution techniques.
    Type: Application
    Filed: December 28, 2017
    Publication date: May 3, 2018
    Inventors: Zehan Wang, Wenzhe Shi, Ferenc Huszar, Robert David Bishop
  • Publication number: 20180075581
    Abstract: A neural network is trained to process received visual data to estimate a high-resolution version of the visual data using a training dataset and reference dataset. A set of training data is generated and a generator convolutional neural network parameterized by first weights and biases is trained by comparing characteristics of the training data to characteristics of the reference dataset. The first network is trained to generate super-resolved image data from low-resolution image data and the training includes modifying first weights and biases to optimize processed visual data based on the comparison between the characteristics of the training data and the characteristics of the reference dataset.
    Type: Application
    Filed: September 15, 2017
    Publication date: March 15, 2018
    Inventors: Wenzhe Shi, Christian Ledig, Zehan Wang, Lucas Theis, Ferenc Huszar