Patents by Inventor Ming-Yu Liu

Ming-Yu Liu 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: 20210157369
    Abstract: An electronic device includes a first body, a second body, two hinges, and at least one electronic assembly. The two hinges are connected between the first body and the second body, and the first body and the second body are adapted to rotate relatively through the two hinges. The electronic assembly is connected to the second body and is located between the two hinges.
    Type: Application
    Filed: February 4, 2021
    Publication date: May 27, 2021
    Applicant: COMPAL ELECTRONICS, INC.
    Inventors: Ming-Chung Peng, Ko-Fan Chen, Chun-Yi Ho, Chien-Ting Lin, Yu-Jung Liu, Hsin-Jung Lee, Hsin-Yu Huang, Jih-Houng Lee, Ming-Feng Liu, Kuo-Jung Wu, Kuo-Pin Chen, Chia-Ling Lee, Jing-Jie Lin
  • Patent number: 11017556
    Abstract: Iterative prediction systems and methods for the task of action detection process an inputted sequence of video frames to generate an output of both action tubes and respective action labels, wherein the action tubes comprise a sequence of bounding boxes on each video frame. An iterative predictor processes large offsets between the bounding boxes and the ground-truth.
    Type: Grant
    Filed: October 4, 2018
    Date of Patent: May 25, 2021
    Assignee: NVIDIA Corporation
    Inventors: Xiaodong Yang, Xitong Yang, Fanyi Xiao, Ming-Yu Liu, Jan Kautz
  • Publication number: 20210150354
    Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.
    Type: Application
    Filed: January 7, 2021
    Publication date: May 20, 2021
    Inventors: Tero Tapani Karras, Samuli Matias Laine, David Patrick Luebke, Jaakko T. Lehtinen, Miika Samuli Aittala, Timo Oskari Aila, Ming-Yu Liu, Arun Mohanray Mallya, Ting-Chun Wang
  • Publication number: 20210150187
    Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.
    Type: Application
    Filed: January 7, 2021
    Publication date: May 20, 2021
    Inventors: Tero Tapani Karras, Samuli Matias Laine, David Patrick Luebke, Jaakko T. Lehtinen, Miika Samuli Aittala, Timo Oskari Aila, Ming-Yu Liu, Arun Mohanray Mallya, Ting-Chun Wang
  • Patent number: 11009796
    Abstract: A method for forming a semiconductor device structure is provided. The method includes forming a material layer over a substrate and forming a resist layer over the material layer. The resist layer includes an inorganic material and an auxiliary. The inorganic material includes a plurality of metallic cores and a plurality of first linkers bonded to the metallic cores. The method includes exposing a portion of the resist layer. The resist layer includes an exposed region and an unexposed region. In the exposed region, the auxiliary reacts with the first linkers. The method also includes removing the unexposed region of the resist layer by using a developer to form a patterned resist layer. The developer includes a ketone-based solvent having a formula (a) or the ester-based solvent having a formula (b).
    Type: Grant
    Filed: September 16, 2019
    Date of Patent: May 18, 2021
    Assignee: Taiwan Semiconductor Manufacturing Co., Ltd.
    Inventors: Ming-Hui Weng, An-Ren Zi, Ching-Yu Chang, Chin-Hsiang Lin, Chen-Yu Liu
  • Patent number: 11011610
    Abstract: A semiconductor device and method for forming the semiconductor device are provided. In some embodiments, a semiconductor substrate comprises a device region. An isolation structure extends laterally in a closed path to demarcate the device region. A first source/drain region and a second source/drain region are in the device region and laterally spaced. A sidewall of the first source/drain region directly contacts the isolation structure at a first isolation structure sidewall, and remaining sidewalls of the first source/drain region are spaced from the isolation structure. A selectively-conductive channel is in the device region, and extends laterally from the first source/drain region to the second source/drain region. A plate comprises a central portion and a first peripheral portion. The central portion overlies the selectively-conductive channel, and the first peripheral portion protrudes from the central portion towards the first isolation structure sidewall.
    Type: Grant
    Filed: April 1, 2020
    Date of Patent: May 18, 2021
    Assignee: Taiwan Semiconductor Manufacturing Co., Ltd.
    Inventors: Chih-Chang Cheng, Fu-Yu Chu, Ming-Ta Lei, Ruey-Hsin Liu, Shih-Fen Huang
  • Publication number: 20210130316
    Abstract: Novel antagonists of toll-like receptor 4 (TLR-4) are provided. More specifically, the novel antagonists of TLR-4 are derived from morphinan. Further, use of said morphinan derivatives in the treatment of diseases and/or disorders mediated by TLR-4, such as autoimmune diseases, inflammation disease and infectious diseases, is provided. Pharmaceutical compositions including said morphinan derivatives are also provided.
    Type: Application
    Filed: December 19, 2017
    Publication date: May 6, 2021
    Applicant: TAIWANJ PHARMACEUTICALS CO., LTD
    Inventors: Syaulan S. YANG, Kuang-Yuan LEE, Edwin SC WU, Ming-Yu HSIAO, Hsiao-Chun WANG, Meng-Hsien LIU, Peter JS CHIU
  • Publication number: 20210134589
    Abstract: A method of forming a pattern in a photoresist includes forming a photoresist layer over a substrate, and selectively exposing the photoresist layer to actinic radiation to form a latent pattern. The latent pattern is developed by applying a developer composition to the selectively exposed photoresist layer to form a pattern. The developer composition includes a first solvent having Hansen solubility parameters of 15<?d<25, 10<?p<25, and 6<?h<30; an acid having an acid dissociation constant, pKa, of ?15<pKa<5, or a base having a pKa of 40>pKa>9.5; and a second solvent having a dielectric constant greater than 18. The first solvent and the second solvent are different solvents.
    Type: Application
    Filed: September 11, 2020
    Publication date: May 6, 2021
    Inventors: Ming-Hui WENG, An-Ren ZI, Ching-Yu CHANG, Chen-Yu LIU
  • Publication number: 20210125036
    Abstract: Apparatuses, systems, and techniques to determine orientation of an objects in an image. In at least one embodiment, images are processed using a neural network trained to determine orientation of an object.
    Type: Application
    Filed: October 29, 2019
    Publication date: April 29, 2021
    Inventors: Jonathan Tremblay, Ming-Yu Liu, Dieter Fox, Philip Ammirato
  • Patent number: 10984286
    Abstract: A style transfer neural network may be used to generate stylized synthetic images, where real images provide the style (e.g., seasons, weather, lighting) for transfer to synthetic images. The stylized synthetic images may then be used to train a recognition neural network. In turn, the trained neural network may be used to predict semantic labels for the real images, providing recognition data for the real images. Finally, the real training dataset (real images and predicted recognition data) and the synthetic training dataset are used by the style transfer neural network to generate stylized synthetic images. The training of the neural network, prediction of recognition data for the real images, and stylizing of the synthetic images may be repeated for a number of iterations. The stylization operation more closely aligns a covariate of the synthetic images to the covariate of the real images, improving accuracy of the recognition neural network.
    Type: Grant
    Filed: February 1, 2019
    Date of Patent: April 20, 2021
    Assignee: NVIDIA Corporation
    Inventors: Aysegul Dundar, Ming-Yu Liu, Ting-Chun Wang, John Zedlewski, Jan Kautz
  • Publication number: 20210097691
    Abstract: Apparatuses, systems, and techniques are presented to generate or manipulate digital images. In at least one embodiment, a network is trained to generate modified images including user-selected features.
    Type: Application
    Filed: September 30, 2019
    Publication date: April 1, 2021
    Inventor: Ming-Yu Liu
  • Publication number: 20210073612
    Abstract: In at least one embodiment, differentiable neural architecture search and reinforcement learning are combined under one framework to discover network architectures with desired properties such as high accuracy, low latency, or both. In at least one embodiment, an objective function for search based on generalization error prevents the selection of architectures prone to overfitting.
    Type: Application
    Filed: September 10, 2019
    Publication date: March 11, 2021
    Inventors: Arash Vahdat, Arun Mohanray Mallya, Ming-Yu Liu, Jan Kautz
  • Publication number: 20210049468
    Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.
    Type: Application
    Filed: October 13, 2020
    Publication date: February 18, 2021
    Inventors: Tero Tapani Karras, Samuli Matias Laine, David Patrick Luebke, Jaakko T. Lehtinen, Miika Samuli Aittala, Timo Oskari Aila, Ming-Yu Liu, Arun Mohanray Mallya, Ting-Chun Wang
  • Patent number: 10922793
    Abstract: Missing image content is generated using a neural network. In an embodiment, a high resolution image and associated high resolution semantic label map are generated from a low resolution image and associated low resolution semantic label map. The input image/map pair (low resolution image and associated low resolution semantic label map) lacks detail and is therefore missing content. Rather than simply enhancing the input image/map pair, data missing in the input image/map pair is improvised or hallucinated by a neural network, creating plausible content while maintaining spatio-temporal consistency. Missing content is hallucinated to generate a detailed zoomed in portion of an image. Missing content is hallucinated to generate different variations of an image, such as different seasons or weather conditions for a driving video.
    Type: Grant
    Filed: March 14, 2019
    Date of Patent: February 16, 2021
    Assignee: NVIDIA Corporation
    Inventors: Seung-Hwan Baek, Kihwan Kim, Jinwei Gu, Orazio Gallo, Alejandro Jose Troccoli, Ming-Yu Liu, Jan Kautz
  • Publication number: 20210042503
    Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.
    Type: Application
    Filed: October 13, 2020
    Publication date: February 11, 2021
    Inventors: Tero Tapani Karras, Samuli Matias Laine, David Patrick Luebke, Jaakko T. Lehtinen, Miika Samuli Aittala, Timo Oskari Aila, Ming-Yu Liu, Arun Mohanray Mallya, Ting-Chun Wang
  • Publication number: 20200410322
    Abstract: Systems and methods that use at least one neural network to infer content of individual frames in a sequence of images and to further infer changes to content in sequence of images over time to determine whether one or more anomalous events are present in sequence of images is described herein.
    Type: Application
    Filed: June 26, 2019
    Publication date: December 31, 2020
    Inventors: Milind Naphade, Tingting Huang, Shuo Wang, Xiaodong Yang, Ming-Yu Liu
  • Patent number: 10872399
    Abstract: Photorealistic image stylization concerns transferring style of a reference photo to a content photo with the constraint that the stylized photo should remain photorealistic. Examples of styles include seasons (summer, winter, etc.), weather (sunny, rainy, foggy, etc.), lighting (daytime, nighttime, etc.). A photorealistic image stylization process includes a stylization step and a smoothing step. The stylization step transfers the style of the reference photo to the content photo. A photo style transfer neural network model receives a photorealistic content image and a photorealistic style image and generates an intermediate stylized photorealistic image that includes the content of the content image modified according to the style image. A smoothing function receives the intermediate stylized photorealistic image and pixel similarity data and generates the stylized photorealistic image, ensuring spatially consistent stylizations.
    Type: Grant
    Filed: January 11, 2019
    Date of Patent: December 22, 2020
    Assignee: NVIDIA Corporation
    Inventors: Yijun Li, Ming-Yu Liu, Ming-Hsuan Yang, Jan Kautz
  • Publication number: 20200394458
    Abstract: Apparatuses, systems, and techniques to detect object in images including digital representations of those objects. In at least one embodiment, one or more objects are detected in an image based, at least in part, on one or more pseudo-labels corresponding to said one or more objects.
    Type: Application
    Filed: June 17, 2019
    Publication date: December 17, 2020
    Inventors: Zhiding Yu, Jason Ren, Xiaodong Yang, Ming-Yu Liu, Jan Kautz
  • Publication number: 20200364303
    Abstract: Apparatuses, systems, and techniques to transfer grammar between sentences. In at least one embodiment, one or more first sentences are translated into one or more second sentences having different grammar using one or more neural networks.
    Type: Application
    Filed: May 15, 2019
    Publication date: November 19, 2020
    Inventors: Ming-Yu Liu, Kevin Lin
  • Publication number: 20200334502
    Abstract: Segmentation is the identification of separate objects within an image. An example is identification of a pedestrian passing in front of a car, where the pedestrian is a first object and the car is a second object. Superpixel segmentation is the identification of regions of pixels within an object that have similar properties. An example is identification of pixel regions having a similar color, such as different articles of clothing worn by the pedestrian and different components of the car. A pixel affinity neural network (PAN) model is trained to generate pixel affinity maps for superpixel segmentation. The pixel affinity map defines the similarity of two points in space. In an embodiment, the pixel affinity map indicates a horizontal affinity and vertical affinity for each pixel in the image. The pixel affinity map is processed to identify the superpixels.
    Type: Application
    Filed: July 6, 2020
    Publication date: October 22, 2020
    Inventors: Wei-Chih Tu, Ming-Yu Liu, Varun Jampani, Deqing Sun, Ming-Hsuan Yang, Jan Kautz