Patents by Inventor Ming Yu

Ming Yu 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: 20230153949
    Abstract: Apparatuses, systems, and techniques are presented to generate one or more images. In at least one embodiment, one or more neural networks are used to generate one or more images based, at least in part, on one or noise values.
    Type: Application
    Filed: November 12, 2021
    Publication date: May 18, 2023
    Inventors: Xun Huang, Zinan Lin, Ming-Yu Liu
  • Publication number: 20230151016
    Abstract: The subject matter described herein is directed to pyruvate kinase activating compounds of Formula I and pharmaceutical salts thereof, methods of preparing the compounds, pharmaceutical compositions comprising the compounds and methods of administering the compounds for the treatment of diseases associated with PKR and/or PKM2, such as pyruvate kinase deficiency, sickle cell disease, and beta-thalassemia.
    Type: Application
    Filed: January 17, 2023
    Publication date: May 18, 2023
    Inventors: Lina Q. Setti, Shahul Nilar, Zhe Li, Ming Yu, Manuel Zancanella
  • Publication number: 20230155301
    Abstract: An antenna module includes a feeding end, multiple first forked radiators, and multiple connecting parts. The first forked radiators are disposed side by side. The connecting parts respectively extend from the feeding end to the first forked radiators. The feeding end, the first forked radiators, and the connecting parts are located on a same plane. The antenna module resonates at a frequency band, and a path length from the feeding end to an end of each of the forked radiators through the corresponding connecting part is ¼ wavelength of the frequency band.
    Type: Application
    Filed: September 8, 2022
    Publication date: May 18, 2023
    Applicant: PEGATRON CORPORATION
    Inventors: Chien-Yi Wu, Wu-Hua Chen, I-Shu Lee, Hung-Ming Yu, Chao-Hsu Wu, Yung-Yi Lee, Man-Jung Tsao, Chi-Min Tang, Shao-Chi Wang
  • Publication number: 20230147641
    Abstract: Apparatuses, systems, and techniques are presented to reconstruct one or more images. In at least one embodiment, one or more neural networks are used to generate one or more images of one or more objects based, at least in part, on input indicating motion of the one or more objects.
    Type: Application
    Filed: November 9, 2021
    Publication date: May 11, 2023
    Inventors: Ting-Chun Wang, Tim Brooks, Ming-Yu Liu, Tero Karras, Jaakko Lehtinen
  • Patent number: 11644726
    Abstract: A display device includes two adjacent pixel electrodes spaced apart from each other by a gap extending in a first direction, at least one signal line extending in the first direction and having at least one protrusion at at least one side thereof, and two adjacent common electrode lines spaced apart from each other. The orthogonal projection of one/the other common electrode line on the substrate is located between the orthogonal projection of the signal line on the substrate and the orthogonal projection of one/the other pixel electrode on the substrate. Each common electrode line has a bend segment bending away from the signal line, wherein the protrusion of the at least one signal line positionally corresponds to the bend segment of each common electrode line, and the length of the protrusion is not larger than the length of the bend segment.
    Type: Grant
    Filed: March 24, 2022
    Date of Patent: May 9, 2023
    Assignee: Au Optronics Corporation
    Inventors: Hung-Che Lin, Yi-Chu Wang, Chien-Huang Liao, Yi-Tse Lin, Fu-ming Yu
  • Publication number: 20230136759
    Abstract: A display device includes two adjacent pixel electrodes spaced apart from each other by a gap extending in a first direction, at least one signal line extending in the first direction and having at least one protrusion at at least one side thereof, and two adjacent common electrode lines spaced apart from each other. The orthogonal projection of one/the other common electrode line on the substrate is located between the orthogonal projection of the signal line on the substrate and the orthogonal projection of one/the other pixel electrode on the substrate. Each common electrode line has a bend segment bending away from the signal line, wherein the protrusion of the at least one signal line positionally corresponds to the bend segment of each common electrode line, and the length of the protrusion is not larger than the length of the bend segment.
    Type: Application
    Filed: March 24, 2022
    Publication date: May 4, 2023
    Applicant: Au Optronics Corporation
    Inventors: Hung-Che Lin, Yi-Chu Wang, Chien-Huang Liao, Yi-Tse Lin, Fu-ming Yu
  • Publication number: 20230128186
    Abstract: An antenna system for improved satellite communication with a ground-based terminal device includes a first antenna, a feeding point, and a phase modulation unit. The first antenna is on a surface of a back cover of the terminal device, and the first antenna comprises a plurality of radiation units in an array. The feeding point feeds power and signals to the first antenna. The phase modulation unit can adjust the transmission phase of the different radiation units within the first antenna. The present disclosure also provides a wireless terminal.
    Type: Application
    Filed: March 4, 2022
    Publication date: April 27, 2023
    Inventors: MING-YU CHOU, CHIA-MING LIANG
  • Publication number: 20230118614
    Abstract: An electronic device and a method for training a neural network model are provided. The method includes: obtaining a first neural network model and a first pseudo-labeled data; inputting the first pseudo-labeled data into the first neural network model to obtain a second pseudo-labeled data; determining whether a second pseudo-label corresponding to the second pseudo-labeled data matching a first pseudo-label corresponding to the first pseudo-labeled data; in response to the second pseudo-label matching the first pseudo-label, adding the second pseudo-labeled data to a pseudo-labeled dataset; and training the first neural network model according to the pseudo-labeled dataset.
    Type: Application
    Filed: November 23, 2021
    Publication date: April 20, 2023
    Applicant: Industrial Technology Research Institute
    Inventors: Mao-Yu Huang, Sen-Chia Chang, Ming-Yu Shih, Tsann-Tay Tang, Chih-Neng Liu
  • Patent number: 11630609
    Abstract: Systems and methods are disclosed for scheduling access commands for a data storage device. A data storage device determines a layout of a plurality of non-volatile memory arrays. The data storage device also determine completed access statistics and pending access statistics for a first set of the plurality of non-volatile memory arrays during a monitoring period. The data storage device further generates a schedule based on the layout of the plurality of non-volatile memory arrays, the completed access statistics, and the pending access statistics and executes access commands based on schedule.
    Type: Grant
    Filed: July 23, 2020
    Date of Patent: April 18, 2023
    Assignee: Western Digital Technologies, Inc.
    Inventors: Haining Liu, YungLi Ji, Yun-Tzuo Lai, Ming-Yu Tai
  • Patent number: 11628388
    Abstract: A filter housing is adapted to be assembled with a cap and installed with an O-ring. The filter housing includes a surrounding wall that is formed with a retaining groove for installation of the O-ring therein, and that has a plurality of protrusions. The retaining groove is defined among inner and outer surrounding surfaces and a groove end surface. The protrusions protrude from one of the inner and outer surrounding surfaces toward the other. The one of the inner and outer surrounding surfaces is adapted to be separated from the O-ring when the surrounding wall is separated from the cap, and is adapted to contact the O-ring when the surrounding wall is assembled with the cap and when the O-ring is squeezed by the cap to deform.
    Type: Grant
    Filed: June 16, 2021
    Date of Patent: April 18, 2023
    Assignee: PA.E Machinery Industrial Co., Ltd.
    Inventor: Ming-Yu Lai
  • Patent number: 11631239
    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: April 22, 2021
    Date of Patent: April 18, 2023
    Assignee: NVIDIA CORPORATION
    Inventors: Xiaodong Yang, Ming-Yu Liu, Jan Kautz, Fanyi Xiao, Xitong Yang
  • Publication number: 20230110206
    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: December 12, 2022
    Publication date: April 13, 2023
    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: 11625613
    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: Grant
    Filed: January 7, 2021
    Date of Patent: April 11, 2023
    Assignee: NVIDIA Corporation
    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: 11621344
    Abstract: A device includes a semiconductor fin, a first epitaxy structure and a gate stack. The semiconductor fin protrudes from a substrate. The first epitaxy feature laterally surrounds a first portion of the semiconductor fin. The gate stack laterally surrounds a second portion of the semiconductor fin above the first portion of the semiconductor fin, wherein the second portion of the semiconductor fin has a lower surface roughness than the first epitaxy feature.
    Type: Grant
    Filed: October 7, 2020
    Date of Patent: April 4, 2023
    Assignee: TAIWAN SEMICONDUCTOR MANUFACTURING CO., LTD.
    Inventors: Wei-Sheng Yun, Shao-Ming Yu, Tung-Ying Lee, Chih-Chieh Yeh
  • Patent number: 11610115
    Abstract: In various examples, a generative model is used to synthesize datasets for use in training a downstream machine learning model to perform an associated task. The synthesized datasets may be generated by sampling a scene graph from a scene grammar—such as a probabilistic grammar—and applying the scene graph to the generative model to compute updated scene graphs more representative of object attribute distributions of real-world datasets. The downstream machine learning model may be validated against a real-world validation dataset, and the performance of the model on the real-world validation dataset may be used as an additional factor in further training or fine-tuning the generative model for generating the synthesized datasets specific to the task of the downstream machine learning model.
    Type: Grant
    Filed: November 15, 2019
    Date of Patent: March 21, 2023
    Assignee: NVIDIA Corporation
    Inventors: Amlan Kar, Aayush Prakash, Ming-Yu Liu, David Jesus Acuna Marrero, Antonio Torralba Barriuso, Sanja Fidler
  • Patent number: 11610122
    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: Grant
    Filed: January 7, 2021
    Date of Patent: March 21, 2023
    Assignee: NVIDIA Corporation
    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: 11611692
    Abstract: A method for identifying and correcting fixed pattern noise includes capturing a focused image and an unfocused image via a variable focus lens. Fixed pattern noise represented in the unfocused image is filtered from the focused image. The unfocused image represents a low-pass filtered component of the focused image; subtracting the unfocused image from the focused image results in a high-pass and fixed pattern noise filtered focused image. Image capture and focus of the variable focus lens are synchronized to remove transitional frames from the image stream.
    Type: Grant
    Filed: November 9, 2020
    Date of Patent: March 21, 2023
    Assignee: Rockwell Collins, Inc.
    Inventors: Christopher A. Keith, Ming Yu
  • Patent number: 11610435
    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: Grant
    Filed: October 13, 2020
    Date of Patent: March 21, 2023
    Assignee: NVIDIA Corporation
    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: 11610290
    Abstract: A system and method. The system may include a display, a lens having distortion, an image generator, and a processor. The lens may be configured to focus light received from an environment. The image generator may be configured to receive the light from the lens and output a stream of images as image data, wherein each of the stream of images is distorted. The processor may be configured to: receive the image data from the image generator; detect a point source object in the stream of images of the image data; enhance the point source object in the stream of images of the image data; undistort the stream of images of the image data having an enhanced point source object; and output a stream of undistorted images as undistorted image data to the display.
    Type: Grant
    Filed: September 24, 2019
    Date of Patent: March 21, 2023
    Assignee: Rockwell Collins, Inc.
    Inventors: Ming Yu, Christopher A. Keith, Gary B. Kingsley
  • Patent number: 11603279
    Abstract: A package structure of a roll-shaped thin film includes a thin film roll and a buffer layer. The thin film roll includes a winding core and a thin film wound around the winding core, and the buffer layer is wound around an outer circumference of the thin film. An end portion of the buffer layer is covered with an inner surface of an end portion of the thin film, and the thickness of the buffer layer is in a range of 1-20 mm.
    Type: Grant
    Filed: September 16, 2021
    Date of Patent: March 14, 2023
    Assignee: CHANG CHUN PETROCHEMICAL CO., LTD.
    Inventors: Ming-Yu Yang, Yao-Sheng Lai, Jui-Chang Chou