Patents Examined by Stephen Smith
  • Patent number: 8149914
    Abstract: A video signal is encoded at a plurality of coding layers exhibiting different spatial resolutions. An input video signal is spatially scaled down into a resolution-lowered video signal that exhibits a resolution lower than the video signal. The resolution-lowered video signal is encoded with a decoding procedure to obtain first coded data and a decoded signal. The decoded signal is spatially scaled up through a high-resolution procedure with reference to the input video signal so that the decoded signal has a smaller error with respect to the input video signal, to obtain a high-resolution scaled-up video signal. The input video signal is encoded through inter-spatial resolution prediction using the high-resolution scaled-up video signal as a predictive signal, to obtain second coded data that exhibits a resolution higher than the resolution-lowered video signal.
    Type: Grant
    Filed: December 20, 2006
    Date of Patent: April 3, 2012
    Assignee: Victor Company of Japan, Ltd.,
    Inventors: Kazuhiro Shimauchi, Satoru Sakazume, Toru Kumakura, Motoharu Ueda
  • Patent number: 8130840
    Abstract: An interpolated image generating apparatus calculates a motion vector reliability level indicating reliability of a motion vector used for inserting an interpolated image, the value of the motion vector reliability level being determined in such a manner that the stronger the correlation is between a source image area and a destination image area that are brought into correspondence with each other by the motion vector, the larger is the value. The apparatus also calculates a failure-preventing vector reliability level indicating reliability of a predetermined failure-preventing vector used for preventing image failures.
    Type: Grant
    Filed: March 20, 2009
    Date of Patent: March 6, 2012
    Assignee: Kabushiki Kaisha Toshiba
    Inventors: Nao Mishima, Ryusuke Hirai, Yasutoyo Takeyama, Goh Itoh
  • Patent number: 8116378
    Abstract: The present invention is to provide a variable length decoding method for decoding complete binary tree code, which is implemented to an entropy coding module for executing the process comprising the steps of: procuring a TabIndex to calculate a value T=?log2(TabIndex)?; reading T bits from a bitstream to obtain a first result M; determining whether or not the result M is smaller than (TabIndex-(1<<T)); if not, obtaining Index equal to (1<<T)?M?1; otherwise, reading 1 bits from the bitstream to obtain a second result N; and then obtaining Index equal to TabIndex-2×M?N?1, so as to decode data stream of video more efficiently and fast.
    Type: Grant
    Filed: December 15, 2006
    Date of Patent: February 14, 2012
    Assignee: ArcSoft, Inc.
    Inventors: Hong-Bo Zhu, Congxiu Wang
  • Patent number: 8077776
    Abstract: Motion estimation is described. A first portion of a predicted frame is obtained. The first portion is for a first predicted value. A first subset of a reference frame is obtained. The first subset is for a first reference value. Twice the first predicted value is subtracted from the first reference value. The outcome of the subtracting is multiplied by the first reference value to produce a partial result. The partial result is used for indication of a degree of difference between the first portion and the first subset.
    Type: Grant
    Filed: December 15, 2006
    Date of Patent: December 13, 2011
    Assignee: Xilinx, Inc.
    Inventor: Mankit Lo