Patents Assigned to Tencent Technology (Shenzhen) Company Limited
  • Patent number: 11967312
    Abstract: This application provides a method for training a semantic understanding model, including: obtaining a first training sample set; performing denoising processing on the first training sample set to form a corresponding second training sample set; processing the second training sample set by using a semantic understanding model, to determine initial parameters of the semantic understanding model; processing the second training sample set by using the semantic understanding model in response to the initial parameters of the semantic understanding model, to determine update parameters of the semantic understanding model; and iteratively updating a semantic representation layer network parameter and a task-related output layer network parameter of the semantic understanding model by using the second training sample set and according to the update parameters of the semantic understanding model.
    Type: Grant
    Filed: October 15, 2021
    Date of Patent: April 23, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Gang Yuan, Xuemin Zhao
  • Patent number: 11967414
    Abstract: This application relates to an image recognition model training method, an image recognition method, apparatus, and system. The method includes: obtaining a to-be-recognized image; extracting image feature information of the to-be-recognized image; and obtaining a lesion category recognition result of the to-be-recognized image by using the image feature information of the to-be-recognized image as an input parameter of a preset image recognition model, the image recognition model being trained by using a training image sample set comprising at least one strong-label training image sample, to determine the lesion category recognition result; and the strong-label training image sample representing an image sample having strong-label information, and the strong-label information comprising at least annotation information of a lesion category and a lesion position in the strong-label training image sample.
    Type: Grant
    Filed: May 14, 2021
    Date of Patent: April 23, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Han Zheng, Zhongqian Sun, Hong Shang, Xinghui Fu, Wei Yang
  • Patent number: 11968366
    Abstract: A video encoding method and apparatus is provided. The method includes determining a chroma format of a video frame, determining, for a currently decoded first quantization matrix (QM), a size of the first QM according to the determined chroma format and a size of a luma transform block (TB) when the first QM is a chroma QM, and decoding the first QM when the size of the first QM is greater than zero. Thus, a size of a chroma QM is flexibly determined according to a chroma format and a size of a luma TB, rather than a fixed size predefined according to an identifier of the QM, thereby improving flexibility during encoding and decoding of the chroma QM. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also provided.
    Type: Grant
    Filed: October 28, 2021
    Date of Patent: April 23, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Hongbin Zhang, Xiang Li, Guichun Li, Shan Liu
  • Patent number: 11966424
    Abstract: Embodiments of this application disclose a method and for dividing geographical regions performed by a computing device. The method includes: encoding latitudes and longitudes in a first dataset to obtain a second dataset, the first dataset recording N groups of latitudes and longitudes, and the second dataset recording N region numbers, and each of the region numbers being used for representing a corresponding geographical region; obtaining a third dataset from the second dataset, the third dataset recording M groups of the region numbers, corresponding central point latitudes and central point longitudes in the geographical regions represented by the region numbers; and dividing the geographical regions represented by the M groups of region numbers into P regions according to the third dataset, each of the regions including at least one of the geographical regions represented by the corresponding group of region numbers.
    Type: Grant
    Filed: May 13, 2021
    Date of Patent: April 23, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Fei Huang, Fan Yang, Yanchun Lin, Jia Liu, Rui Guo
  • Patent number: 11968427
    Abstract: This application provides a video message generation method and apparatus, an electronic device, and a storage medium. The video message generation method includes: receiving a comment instruction for a video; determining a target frame corresponding to the comment instruction in the video; obtaining input comment content of the target frame; and generating a video message based on the video, the target frame, and the comment content of the target frame.
    Type: Grant
    Filed: March 29, 2022
    Date of Patent: April 23, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Fen He, Xuejun Sun, Han Liu, Liqiang Liu, Yujun Liu
  • Patent number: 11968379
    Abstract: Disclosed is a video decoding method, including: obtaining a current video frame, the current video frame being partitioned into a plurality of regions; obtaining a syntax element carried in syntax data corresponding to each of the plurality of regions, the syntax element being used for indicating a resolution used to decode the region, and a plurality of resolutions used to decode the plurality of regions including at least two different resolutions; and decoding the each of the plurality of regions by using the resolution corresponding to the region. The plurality of resolutions are determined according to a transmission bandwidth of a video stream including the current video frame from a source to a destination, e.g., by comparing the transmission bandwidth with a preset bandwidth threshold.
    Type: Grant
    Filed: September 8, 2021
    Date of Patent: April 23, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Xinwei Gao, Weiran Li, Chenchen Gu
  • Patent number: 11968294
    Abstract: This application provide a data management method for a blockchain system, a medium, and an electronic device. The system includes an accounting node sub-network and a service node sub-network. The method includes: adding, after an accounting node generates a first data block, first key information used for verifying a block header of a second data block generated after the first data block to a block header of the first data block; generating a signature corresponding to the first data block, and adding the signature corresponding to the first data block to the block header of the first data block; and releasing the block header of the first data block to the service node sub-network, to cause a service node to verify the signature included in the block header of the first data block, and obtaining the first key information after a successful verification to verify the block header of the second data block.
    Type: Grant
    Filed: January 13, 2021
    Date of Patent: April 23, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Maocai Li, Zongyou Wang, Li Kong, Kaiban Zhou, Hu Lan, Yifang Shi, Changqing Yang, Jinsong Zhang, Yong Ding, Gengliang Zhu, Qucheng Liu, Qiuping Chen
  • Publication number: 20240129143
    Abstract: A multi-blockchain processing method includes receiving a first service request from a first service node, and invoking a first cross-chain reading contract on a first chain to read first service associated information from a target chain. The method further includes invoking, by a first consensus node, a first service processing contract to perform the first service and obtain service data as a first service execution result, and writing the first service execution result to the first chain. The method further includes reading, by the first consensus node, the service data from the first chain based on a second service indicated in a cross-chain reading request, and returning the service data to a second consensus node configured to write a second service execution result generated by the second service to the second chain after performing the second service based on the service data generated by the first service.
    Type: Application
    Filed: October 12, 2023
    Publication date: April 18, 2024
    Applicant: Tencent Technology (Shenzhen) Company Limited
    Inventor: Gengliang ZHU
  • Patent number: 11961237
    Abstract: Embodiments of this application disclose a foreground data generation method performed at a computer device. The method includes: obtaining a background image and a target image, the target image containing a target object and a background; removing the background from the target image according to the background image and the target image, to obtain initial foreground data of the target object in the target image; obtaining certain foreground data and uncertain data from the initial foreground data, wherein the uncertain data represents data whose value is between the certain foreground data and background data corresponding to the background; and segmenting the certain foreground data from the uncertain data, to obtain target foreground data of the target object in the target image.
    Type: Grant
    Filed: May 25, 2021
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Jingmin Luo, Xiaolong Zhu
  • Patent number: 11961609
    Abstract: The disclosure provides a three-dimensional (3D) image classification method and apparatus, a device, and a storage medium. The method includes: obtaining a 3D image, the 3D image including first-dimensional image information, second-dimensional image information, and third-dimensional image information; extracting a first image feature corresponding to planar image information from the 3D image; extracting a second image feature corresponding to the third-dimensional image information from the 3D image; fusing the first image feature and the second image feature, to obtain a fused image feature corresponding to the 3D image; and determining a classification result corresponding to the 3D image according to the fused image feature corresponding to the 3D image.
    Type: Grant
    Filed: July 23, 2021
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Yi Fan Hu, Ye Feng Zheng
  • Patent number: 11961325
    Abstract: An image processing method and apparatus, a computer-readable medium, and an electronic device are provided. The image processing method includes: respectively projecting, according to a plurality of view angle parameters corresponding to a plurality of view angles, a face model of a target object onto a plurality of face images of the target object acquired from the plurality of view angles, to determine correspondences between regions on the face model and regions on the face image; respectively extracting, based on the correspondences and a target region in the face model that need to generate a texture image, images corresponding to the target region from the plurality of face images; and fusing the images that correspond to the target region and that are respectively extracted from the plurality of face images, to generate the texture image.
    Type: Grant
    Filed: February 24, 2021
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Xiangkai Lin, Linchao Bao, Yonggen Ling, Yibing Song, Wei Liu
  • Patent number: 11961242
    Abstract: A target tracking method is provided for a computer device. The method includes determining a target candidate region of a current image frame; capturing a target candidate image matching the target candidate region from the current image frame; determining a target region of the current image frame according to an image feature of the target candidate image; determining motion prediction data of a next image frame relative to the current image frame by using a motion prediction model and according to the image feature of the target candidate image; and determining a target candidate region of the next image frame according to the target region and the motion prediction data.
    Type: Grant
    Filed: September 25, 2020
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Yandan Zhao, Chengjie Wang, Weijian Cao, Yun Cao, Pan Cheng, Yuan Huang
  • Patent number: 11962548
    Abstract: This disclosure includes a message display method and apparatus, a device, and a storage medium. In the method, a message window of a first message group is displayed. The message window is configured to display messages of the first message group from a plurality of user accounts. A first message in the message window of the first message group is added to a grouped message window. The grouped message window is configured to display messages associated with a topic of the first message. Further, the messages associated with the topic are from at least one message group.
    Type: Grant
    Filed: May 19, 2021
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGIES (SHENZHEN) COMPANY LIMITED
    Inventors: Yuan Tian, Yijie Shen
  • Patent number: 11961233
    Abstract: This application provides a method and apparatus for training an image segmentation model, a device, and a storage medium.
    Type: Grant
    Filed: September 9, 2021
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventor: Luyan Liu
  • Patent number: 11961226
    Abstract: In a medical image recognition method, applied to a computer device, a to-be-recognized medical image set is obtained, where the to-be-recognized medical image set includes at least one to-be-recognized medical image. A to-be-recognized area corresponding to each to-be-recognized medical image in the to-be-recognized medical image set is extracted. The to-be-recognized area is a part of the to-be-recognized medical image. A recognition result of each to-be-recognized area through a medical image recognition model is determined. The medical image recognition model is obtained through training according to a medical image sample set. The medical image sample set includes at least one medical image sample, and each medical image sample carries corresponding annotation information. The annotation information is used for representing a type of the medical image sample, and the recognition result is used for representing the a of the to-be-recognized medical image.
    Type: Grant
    Filed: October 23, 2020
    Date of Patent: April 16, 2024
    Assignee: Tencent Technology (Shenzhen) Company Limited
    Inventors: Kaiwen Xiao, Zhongqian Sun, Chen Cheng, Wei Yang
  • Patent number: 11962930
    Abstract: This application discloses a method for controlling a plurality of virtual characters in an application with an augmented reality (AR) function running in the a terminal. The method includes: displaying a first user interface of the application, the first user interface comprising a plurality of virtual characters for user selection; receiving a first user selection operation on at least two virtual characters on the first user interface; displaying a second user interface of the application, the second user interface including a background picture of a real world and the at least two virtual characters located in the background picture; receiving a second user selection operation on the second user interface; and determining a target virtual character from the at least two virtual characters according to the second user selection operation and the rendering sequence.
    Type: Grant
    Filed: July 2, 2021
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Yizhou Zhuang, Zhiyi Zhou
  • Patent number: 11962897
    Abstract: Some aspects of the disclosure provide a method for camera operation control in an electronic device. The method includes obtaining, based on a plurality of frames associated with a target time slice, a first target parameter to be met by a camera of a virtual camera system in the target time slice. Then, the method includes determining a target operation speed of the camera in the target time slice at least partially based on the first target parameter and a time-speed change magnification curve, and controlling the camera to operate based on the target operation speed. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also contemplated.
    Type: Grant
    Filed: February 18, 2022
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Qi Liu, Guangzhou Zhai, Weihua Cui
  • Patent number: 11961064
    Abstract: An information display method, a storage medium, and an electronic device are provided. The method includes: obtaining, by a terminal, resource transfer information used for transferring a resource; splitting, by the terminal, the resource transfer information into multiple pieces of resource transfer sub information; and displaying, by the terminal, a dynamic screen for indicating the resource transfer sub information. A quantity of pieces of the resource transfer sub information indicated on the dynamic screen at one moment is less than a total quantity of the multiple pieces of resource transfer sub information.
    Type: Grant
    Filed: May 9, 2022
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventor: Haiba Zhang
  • Patent number: 11960571
    Abstract: A method for training an image recognition model includes: obtaining training image sets; obtaining a first predicted probability, a second predicted probability, a third predicted probability, and a fourth predicted probability based on the training image sets by using an initial image recognition model; determining a target loss function according to the first predicted probability, the second predicted probability, the third predicted probability, and the fourth predicted probability; and training the initial image recognition model based on the target loss function, to obtain an image recognition model.
    Type: Grant
    Filed: October 29, 2021
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Hong Shang, Han Zheng, Zhongqian Sun
  • Patent number: 11960973
    Abstract: This application relates to a method for analyzing crosstalk between qubits, performed by a terminal. The method includes identifying a first qubit and a second qubit; performing spectral quantum process tomography on quantum states corresponding to the first qubit and the second qubit, to obtain a first eigenspectrum of a signal function corresponding to the first qubit and a second eigenspectrum of a signal function corresponding to the second qubit; performing spectral quantum process tomography on the quantum states corresponding to the first qubit and the second qubit, to obtain a third eigenspectrum of a common signal function of the first qubit and the second qubit; and determining a crosstalk intensity between the first qubit and the second qubit based on the first eigenspectrum, the second eigenspectrum, and the third eigenspectrum.
    Type: Grant
    Filed: February 17, 2022
    Date of Patent: April 16, 2024
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Yuqin Chen, Shengyu Zhang