Patents by Inventor Megamus Zhang

Megamus Zhang 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: 11430134
    Abstract: An optical flow accelerator (OFA) which provides hardware-based acceleration of optical flow and stereo disparity determination is described. A system is described which includes an OFA configured to determine a first optical flow using a first disparity search technique, and to determine a second optical flow using a second disparity search technique that is different from the first disparity search technique. The system also includes a processor configured to combine the first optical flow and the second optical flow to generate a third optical flow. In some implementations, the first and second disparity search techniques are based upon Semi-Global Matching (SGM). In some implementations, the OFA is further configurable to determine stereo disparity.
    Type: Grant
    Filed: September 3, 2019
    Date of Patent: August 30, 2022
    Assignee: NVIDIA CORPORATION
    Inventors: Dong (Megamus) Zhang, JinYue (Gser) Lu, Zejun (Harry) Hu
  • Patent number: 11064203
    Abstract: Real-time, hardware-implementable Structured Similarity (SSIM)-based rate distortion optimization (RDO) techniques for video transmission are described. The disclosed techniques provide efficient application of SSIM as a distortion metric in selecting prediction modes for encoding video for transmission. A prediction mode, at a high level, specifies which previously encoded group of pixels can be utilized to encode a subsequent block of pixels in a video frame. A less compute intensive distortion metric is first used to select a subset of candidate prediction modes. Then a more compute intensive SSIM-based selection is made on the subset. By utilizing the disclosed techniques during video encoding, tradeoffs between distortion and transmission rate can be made that are more relevant to human perception.
    Type: Grant
    Filed: March 12, 2018
    Date of Patent: July 13, 2021
    Assignee: NVIDIA CORPORATION
    Inventors: Megamus Zhang, Jant Chen, Steven Feng, Shining Yi
  • Publication number: 20210065379
    Abstract: An optical flow accelerator (OFA) which provides hardware-based acceleration of optical flow and stereo disparity determination is described. A system is described which includes an OFA configured to determine a first optical flow using a first disparity search technique, and to determine a second optical flow using a second disparity search technique that is different from the first disparity search technique. The system also includes a processor configured to combine the first optical flow and the second optical flow to generate a third optical flow. In some implementations, the first and second disparity search techniques are based upon Semi-Global Matching (SGM). In some implementations, the OFA is further configurable to determine stereo disparity.
    Type: Application
    Filed: September 3, 2019
    Publication date: March 4, 2021
    Inventors: Dong (Megamus) Zhang, JinYue (Gser) Lu, Zejun (Harry) Hu
  • Publication number: 20190281302
    Abstract: Real-time, hardware-implementable Structured Similarity (SSIM)-based rate distortion optimization (RDO) techniques for video transmission are described. The disclosed techniques provide efficient application of SSIM as a distortion metric in selecting prediction modes for encoding video for transmission. A prediction mode, at a high level, specifies which previously encoded group of pixels can be utilized to encode a subsequent block of pixels in a video frame. A less compute intensive distortion metric is first used to select a subset of candidate prediction modes. Then a more compute intensive SSIM-based selection is made on the subset. By utilizing the disclosed techniques during video encoding, tradeoffs between distortion and transmission rate can be made that are more relevant to human perception.
    Type: Application
    Filed: March 12, 2018
    Publication date: September 12, 2019
    Inventors: Megamus Zhang, Jant Chen, Steven Feng, Shining Yi