Patents by Inventor Wei-Cheng Kuo

Wei-Cheng Kuo 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: 20250110504
    Abstract: Aspects of the present disclosure relate to systems and methods for mobile device control using language input.
    Type: Application
    Filed: September 18, 2024
    Publication date: April 3, 2025
    Inventors: Wei Cheng KUO, Hugues THOMAS, Jian ZHANG
  • Patent number: 12243297
    Abstract: A computer-implemented method can include a training phase and a hemorrhage detection phase. The training phase can include: receiving a first plurality of frames from at least one original computed tomography (CT) scan of a target subject, wherein each frame may or may not include a visual indication of a hemorrhage, and further wherein each frame including a visual indication of a hemorrhage has at least one label associated therewith; and using a fully convolutional neural network (FCN) to train a model by determining, for each of the first plurality of frames, whether at least one sub-portion of the frame includes a visual indication of a hemorrhage and classifying the sub-portion of the frame based on the determining.
    Type: Grant
    Filed: July 20, 2020
    Date of Patent: March 4, 2025
    Assignee: THE REGENTS OF THE UNIVERSITY OF CALIFORNIA
    Inventors: Jitendra Malik, Pratik Mukherjee, Esther L. Yuh, Wei-Cheng Kuo, Christian Haene
  • Publication number: 20250005924
    Abstract: Provided are machine-learned models for performing video processing with improved efficiency. In particular, the machine-learned model can perform the sparse application of one or more video kernels to a set of video data to generate video tokens that can, for example, be provided as input to a visual transformer. Thus, example implementations of the present disclosure are directed to an approach which can turn a visual transformer (e.g., a ViT encoder) into an efficient video model. Furthermore, example implementations described herein can seamlessly work with both image and video inputs. Specifically, by sparsely sampling the inputs, the model is able to do training and inference from both inputs. The proposed model is easily scalable and can optionally be adapted to large-scale pre-trained visual transformers without requiring full finetuning.
    Type: Application
    Filed: November 22, 2023
    Publication date: January 2, 2025
    Inventors: Anthony J. Piergiovanni, Wei-Cheng Kuo, Anelia Angelova
  • Patent number: 12093014
    Abstract: A position calibration system and method are disclosed, in which a control unit is provided to control a positioner sensing module to scan a circular positioner provided on a positioning substrate in a first direction and a second direction so as to acquire midpoints of two scanned line segments and acquire an intersection of lines extending from the two center points in a direction perpendicular to the first and the second directions as a calibration reference point, which correspond to a centroid (a center) of the circular positioner. The calibration reference point functions as a reference point for positioning the positioning substrate with respect to the positioner sensing module and is stored in a memory unit. The calibration reference point can be used as a positioning point during installation of a machine and can also be used for calibration of a position of the machine.
    Type: Grant
    Filed: January 21, 2022
    Date of Patent: September 17, 2024
    Assignee: CHROMA ATE INC.
    Inventors: Chin-Yi Ouyang, Wei-Cheng Kuo, Chien-Ming Chen, Xin-Yi Wu
  • Publication number: 20240289981
    Abstract: Generally, the disclosure is directed to generalized objected location, where the located object is in accordance to a natural language (NL) query. More specifically, the embodiments include a unified generalized visual localization architecture. The architecture achieves enhanced performance on the following three tasks: referring expression comprehension, object localization, and object detection. The embodiments employ machine-learned NL models and/or image models. The architecture is enabled to understand and answer natural localization questions towards an image, to output multiple boxes, provide no output if the object is not present (e.g., a null result), as well as, solve general detection tasks.
    Type: Application
    Filed: February 23, 2023
    Publication date: August 29, 2024
    Inventors: Wei-Cheng Kuo, Fred Bertsch, Wei Li, Anthony J. Piergiovanni, Mohammad Taghi Saffar, Anelia Angelova
  • Publication number: 20230394306
    Abstract: Provided is an efficient multi-modal processing model. The multi-modal processing model can process input data from multiple different domains to generate a prediction for a multi-modal processing task. A machine-learned multi-modal processing model can include an adaptive tokenization layer that is configured to adaptively tokenize features generated from the multi-modal inputs into sets of tokens. Specifically, the tokens may have a smaller data size relative to the features from the inputs, thereby enabling a reduced number of processing operations to be performed overall, thereby improving the efficiency of model.
    Type: Application
    Filed: June 2, 2023
    Publication date: December 7, 2023
    Inventors: Anthony J. Piergiovanni, Wei-Cheng Kuo, Anelia Angelova
  • Patent number: 11740283
    Abstract: The present invention relates to a multistory electronic device testing apparatus, which mainly comprises a feeding and binning device, a multi-axis transfer device, a chip-testing device and a main controller. The feeding and binning device includes an upper module and a lower module. The chip-testing device includes a plurality of testing units arranged vertically. The main controller not only controls the feeding, binning and testing operations, but also controls the multi-axis transfer device to transfer an electronic device to be tested or a tested electronic device between the feeding and binning device and the chip-testing device. Accordingly, the three-dimensional arrangement of the feeding and binning module and the testing device is realized, and the accommodating capacity and the testing capacity for the electronic devices to be tested and the tested electronic devices can be increased.
    Type: Grant
    Filed: June 7, 2022
    Date of Patent: August 29, 2023
    Assignee: CHROMA ATE INC.
    Inventors: Chin-Yi Ouyang, Chien-Ming Chen, Wei-Cheng Kuo, Xin-Yi Wu, Iching Tsai
  • Publication number: 20230022501
    Abstract: The present invention relates to a multistory electronic device testing apparatus, which mainly comprises a feeding and binning device, a multi-axis transfer device, a chip-testing device and a main controller. The feeding and binning device includes an upper module and a lower module. The chip-testing device includes a plurality of testing units arranged vertically. The main controller not only controls the feeding, binning and testing operations, but also controls the multi-axis transfer device to transfer an electronic device to be tested or a tested electronic device between the feeding and binning device and the chip-testing device. Accordingly, the three-dimensional arrangement of the feeding and binning module and the testing device is realized, and the accommodating capacity and the testing capacity for the electronic devices to be tested and the tested electronic devices can be increased.
    Type: Application
    Filed: June 7, 2022
    Publication date: January 26, 2023
    Inventors: Chin-Yi OUYANG, Chien-Ming CHEN, Wei-Cheng KUO, Xin-Yi WU, Iching TSAI
  • Publication number: 20230023844
    Abstract: A position calibration system and method are disclosed, in which a control unit is provided to control a positioner sensing module to scan a circular positioner provided on a positioning substrate in a first direction and a second direction so as to acquire midpoints of two scanned line segments and acquire an intersection of lines extending from the two center points in a direction perpendicular to the first and the second directions as a calibration reference point, which correspond to a centroid (a center) of the circular positioner. The calibration reference point functions as a reference point for positioning the positioning substrate with respect to the positioner sensing module and is stored in a memory unit. The calibration reference point can be used as a positioning point during installation of a machine and can also be used for calibration of a position of the machine.
    Type: Application
    Filed: January 21, 2022
    Publication date: January 26, 2023
    Inventors: Chin-Yi Ouyang, Wei-Cheng Kuo, Chien-Ming Chen, Xin-Yi Wu
  • Publication number: 20220254147
    Abstract: A computer-implemented method can include a training phase and a hemorrhage detection phase. The training phase can include: receiving a first plurality of frames from at least one original computed tomography (CT) scan of a target subject, wherein each frame may or may not include a visual indication of a hemorrhage, and further wherein each frame including a visual indication of a hemorrhage has at least one label associated therewith; and using a fully convolutional neural network (FCN) to train a model by determining, for each of the first plurality of frames, whether at least one sub-portion of the frame includes a visual indication of a hemorrhage and classifying the sub-portion of the frame based on the determining.
    Type: Application
    Filed: July 20, 2020
    Publication date: August 11, 2022
    Inventors: JITENDRA MALIK, PRATIK MUKHERJEE, ESTHER L. YUH, WEI-CHENG KUO, CHRISTIAN HAENE
  • Publication number: 20210092940
    Abstract: The present invention provides a uniform poultry feeding system that is comprised of a trough and a fodder conveyor unit. The trough consists of a pair of trough bodies and a fencing. The fodder conveyor unit is installed above the trough and has a conveyor being set between the trough bodies and the fencing, a slot at the bottom to enable fodder falling into the trough bodies, several stopcocks fixed onto the pipe, several suspension cables extending and connecting to the stopcocks to control the rise and fall of the stopcocks and the conveyor by moving up and down, prevent rotation of the pipe and displacement of the slot and adjust the discharge of fodder into the trough bodies by changing the distance between the conveyor and the bottom of the trough bodies.
    Type: Application
    Filed: July 27, 2017
    Publication date: April 1, 2021
    Inventors: Wei-Cheng KUO, Shu-Mei Hung
  • Patent number: 9384537
    Abstract: A method and non-transitory computer readable medium for processing an oversampled image is disclosed. Filters are applied to an oversampled image to obtain a filtered image. The image filters are Gabor filter, cosine filter, laplacian of Gaussian filter, and radial filter. The filtered image can be turned into a displayed image that is displayed. The displayed image can be a 3D image. The displayed image can be refreshed at a rate of about one frame per second.
    Type: Grant
    Filed: August 31, 2014
    Date of Patent: July 5, 2016
    Assignee: NATIONAL TAIWAN UNIVERSITY
    Inventors: Chi-Kuang Sun, Wei-Cheng Kuo
  • Publication number: 20160063689
    Abstract: A method and non-transitory computer readable medium for processing an oversampled image is disclosed. Filters are applied to an oversampled image to obtain a filtered image. The image filters are Gabor filter, cosine filter, laplacian of Gaussian filter, and radial filter. The filtered image can be turned into a displayed image that is displayed. The displayed image can be can be a 3D image. The displayed image can be refreshed at a rate of about one frame per second.
    Type: Application
    Filed: August 31, 2014
    Publication date: March 3, 2016
    Inventors: Chi-Kuang Sun, Wei-Cheng Kuo