中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

32 records (took 0.022 seconds)

    IMAGE ENCODING AND DECODING APPARATUS AND METHOD
    21.
    发明申请
    IMAGE ENCODING AND DECODING APPARATUS AND METHOD 审中-公开

    公开(公告)号：US20150229936A1

    公开(公告)日：2015-08-13

    申请号：US14666951

    申请日：2015-03-24

    Applicant: SAMSUNG ELECTRONICS CO., LTD.

    Inventor： Hyun Mun KIM ,  Dae-sung CHO ,  Woo-shik KIM

    IPC: H04N19/44 ,  H04N19/50

    CPC classification number: H04N19/44 ,  H04N19/105 ,  H04N19/107 ,  H04N19/136 ,  H04N19/142 ,  H04N19/162 ,  H04N19/176 ,  H04N19/182 ,  H04N19/184 ,  H04N19/46 ,  H04N19/50 ,  H04N19/61

    Abstract: An image encoding and decoding apparatus and method, the image encoding and decoding apparatus includes: an encoding unit adjusting at least one weight value adaptively to the number of bits expressing each pixel of a current image, predicting a pixel value of a block of interest by performing an operation on a pixel value of at least one block having a pixel value similar to the pixel value of the block of interest among blocks included in a previous image and the at least one adjusted weight value, and encoding the pixel value of the block of interest using the predicted pixel value of the block of interest; and a decoding unit adjusting the at least one weight value adaptively to the number of bits, restoring the predicted pixel value of the block of interest by performing an operation on the pixel value of the at least one block having the pixel value similar to the pixel value of the block of interest among blocks included in the restored previous image and the at least one adjusted weight value, and decoding the pixel value of the block of interest using the restored pixel value, wherein the previous image refers to an image displayed prior to a current image, and the block of interest is a target block of interest among blocks belonging to the current image.

    METHOD AND APPARATUS FOR ENCODING CLOUD DISPLAY SCREEN BY USING APPLICATION PROGRAMMING INTERFACE INFORMATION
    24.
    发明申请
    METHOD AND APPARATUS FOR ENCODING CLOUD DISPLAY SCREEN BY USING APPLICATION PROGRAMMING INTERFACE INFORMATION 有权
    使用应用编程接口信息编辑云显示屏的方法和装置

    公开(公告)号：US20140099041A1

    公开(公告)日：2014-04-10

    申请号：US14045333

    申请日：2013-10-03

    Applicant: SAMSUNG ELECTRONICS CO., LTD.

    Inventor： Dae-hee KIM ,  Dai-woong CHOI ,  Dae-sung CHO

    IPC: G06T9/00

    CPC classification number: G06T9/00 ,  H04N19/12 ,  H04N19/136 ,  H04N19/167 ,  H04N19/17 ,  H04N19/20 ,  H04N19/85 ,  H04N21/431 ,  H04N21/8166

    Abstract: A method and apparatus for implementing cloud computing by efficiently encoding a display screen between a cloud server and a client terminal. The method for encoding a cloud display screen in the cloud server, includes: acquiring region information of currently generated windows based on an application programming interface (API) function information of an Operating System (OS); extracting relative depth information between each of the currently generated windows from the OS; generating an encoding mode map in which an entire screen is divided into a plurality of blocks, based on the acquired region information and the extracted relative depth information; and encoding each of the plurality of blocks based on the generated encoding mode map.

    Abstract translation: 一种通过在云服务器和客户终端之间高效地编码显示屏来实现云计算的方法和装置。 云服务器中的云显示屏的编码方法包括：基于操作系统（OS）的应用编程接口（API）功能信息获取当前生成的窗口的区域信息; 从OS中提取当前生成的每个窗口之间的相对深度信息; 基于获取的区域信息和所提取的相对深度信息，生成其中整个屏幕被划分为多个块的编码模式图; 以及基于所生成的编码模式图来编码所述多个块中的每一个。

    METHOD AND DEVICE FOR PROCESSING MULTI-CHANNEL FEATURE MAP IMAGES
    26.
    发明申请
    METHOD AND DEVICE FOR PROCESSING MULTI-CHANNEL FEATURE MAP IMAGES 审中-公开

    公开(公告)号：US20180350110A1

    公开(公告)日：2018-12-06

    申请号：US15967039

    申请日：2018-04-30

    Applicant: SAMSUNG ELECTRONICS CO., LTD.

    Inventor： Dae-sung CHO ,  Won-jae LEE

    IPC: G06T9/00 ,  G06K9/62 ,  G06N3/04

    Abstract: A convolutional neural network-based image processing method is provided. The method includes: receiving, in a second layer, multi-channel feature map images generated by applying a convolution operation to an input image of a convolutional neural network having a plurality of layers with a plurality of filter kernels of a first layer; analyzing a dynamic range of the multi-channel feature map images; re-ordering the multi-channel feature map images, based on the dynamic range; and processing the re-ordered multi-channel feature map images in the second layer.

    METHOD FOR RE-ENCODING IMAGE AND APPARATUS THEREFOR
    28.
    发明申请
    METHOD FOR RE-ENCODING IMAGE AND APPARATUS THEREFOR 审中-公开

    公开(公告)号：US20170318291A1

    公开(公告)日：2017-11-02

    申请号：US15517357

    申请日：2014-12-29

    Applicant: SAMSUNG ELECTRONICS CO., LTD.

    Inventor： Dai-woong CHOI ,  Woong-il CHOI ,  Dae-sung CHO

    IPC: H04N19/124 ,  H04N19/184 ,  H04N19/625 ,  H04N19/91

    Abstract: Provided is a re-encoding method including obtaining a first quantization table from a bitstream including an image encoded using the first quantization table; obtaining a second quantization table based on a pattern representing a size distribution of values of elements of the first quantization table, the second quantization table including elements respectively corresponding to the elements of the first quantization table; and re-encoding a reconstructed image by using the second quantization table, the reconstructed image being obtained by decoding the encoded image by using the first quantization table.

Patent Agency Ranking