Motion estimation apparatus and method
    1.
    发明授权
    Motion estimation apparatus and method 有权
    运动估计装置及方法

    公开(公告)号:US08737478B2

    公开(公告)日:2014-05-27

    申请号:US13297291

    申请日:2011-11-16

    IPC分类号: H04N7/12 H04N11/02 H04N7/26

    摘要: Disclosed is a motion estimation apparatus and method capable of reduce a memory bandwidth by accessing reference image data occupying the largest part of the memory bandwidth based on a result of motion estimation skip determination. The motion estimation apparatus includes a motion estimation skip determination unit, a sum absolute of absolute difference (SAD) processing unit and a multiplexer (MUX) unit. The motion estimation skip determination unit determines the presence of motion estimation skip by reading current image data and skip reference image data, and outputs a motion vector in a skip mode. The SAD processing unit calculates an SAD value by reading the current image data and reference image data and outputs a motion vector in a normal mode. The MUX unit outputs the motion vector in the skip mode or the motion vector in the normal mode as a final motion vector based on a skip flag signal.

    摘要翻译: 公开了一种运动估计装置和方法,其能够通过基于运动估计跳跃确定的结果访问占用存储器带宽的最大部分的参考图像数据来减少存储器带宽。 运动估计装置包括运动估计跳过确定单元,绝对差(SAD)处理单元和多路复用器(MUX)单元的和绝对值。 运动估计跳过确定单元通过读取当前图像数据并跳过参考图像数据来确定运动估计跳过的存在,并且以跳过模式输出运动矢量。 SAD处理单元通过读取当前图像数据和参考图像数据来计算SAD值,并以正常模式输出运动矢量。 MUX单元基于跳过标志信号将跳过模式中的运动矢量或正常模式中的运动矢量作为最终运动矢量输出。

    Accessing value for local variable from function call stack upon offset matching with instruction extracted stack pointer offset or from cache
    2.
    发明授权
    Accessing value for local variable from function call stack upon offset matching with instruction extracted stack pointer offset or from cache 有权
    通过与指令提取的堆栈指针偏移或从缓存进行偏移匹配,从函数调用堆栈访问局部变量的值

    公开(公告)号:US08478970B2

    公开(公告)日:2013-07-02

    申请号:US12832313

    申请日:2010-07-08

    IPC分类号: G06F9/312 G06F12/08

    CPC分类号: G06F9/342 G06F9/3824

    摘要: There is provided a method and apparatus for accessing a memory according to a processor instruction. The apparatus includes: a stack offset extractor extracting an offset value from a stack pointer offset indicating a local variable in the processor instruction; a local stack storage including a plurality of items, each of which is formed of an activation bit indicating whether each item is activated, an offset storing an offset value of a stack pointer, and an element storing a local variable value of the stack pointer; an offset comparator comparing the extracted offset value with an offset value of each item and determining whether an item corresponding to the extracted offset value is present in the local stack storage; and a stack access controller controlling a processor to access the local stack storage or a cache memory according to a determining result of the offset comparator.

    摘要翻译: 提供了根据处理器指令访问存储器的方法和装置。 该装置包括:堆栈偏移提取器,从指示处理器指令中的局部变量的堆栈指针偏移量中提取偏移值; 本地堆栈存储器,包括多个项目,每个项目由表示每个项目是否被激活的激活位,存储堆栈指针的偏移值的偏移量和存储堆栈指针的局部变量值的元素组成; 偏移比较器,将所提取的偏移值与每个项目的偏移值进行比较,并确定与所提取的偏移值相对应的项目是否存在于本地堆栈存储器中; 以及堆栈访问控制器,其根据偏移比较器的确定结果控制处理器访问本地堆栈存储器或高速缓冲存储器。

    Method of block partition for H.264 inter prediction
    3.
    发明授权
    Method of block partition for H.264 inter prediction 有权
    H.264帧间预测的块分割方法

    公开(公告)号:US08265155B2

    公开(公告)日:2012-09-11

    申请号:US12565949

    申请日:2009-09-24

    IPC分类号: H04N7/12

    摘要: Provided is a method for partitioning a block in inter prediction including selecting one standard reference frame from at least one reference frame which is a comparison target in inter prediction, searching whether or not a higher-level macroblock of a current frame is partitioned based on the selected standard reference frame, determining a partition size of the higher-level macroblock of the current frame, searching whether or not the higher-level macroblock in which the partition size is determined is partitioned by comparing with any certain reference frame other than the standard reference frame, and stopping a block size search on the higher-level macroblock of the current frame if a partition size of the higher-level macroblock of the current frame determined by comparing with the standard reference frame and a partition size of the higher-level macroblock of the current frame determined by comparing with the certain reference frame are different from each other.

    摘要翻译: 提供了一种用于在帧间预测中划分块的方法,包括从作为帧间预测中的比较对象的至少一个参考帧中选择一个标准参考帧,基于所述帧间预测是否划分当前帧的更高级宏块 选择标准参考帧,确定当前帧的较高级别宏块的分区大小,通过与除标准参考之外的任何特定参考帧进行比较来分割其中确定分区大小的更高级宏块是否被划分 并且如果通过与标准参考帧比较确定的当前帧的较高级宏块的分区大小和较高级宏块的分区大小,则停止当前帧的较高级宏块上的块大小搜索 通过与特定参考系进行比较确定的当前帧不同。

    ADAPTIVE MOTION SEARCH RANGE DETERMINING APPARATUS AND METHOD
    4.
    发明申请
    ADAPTIVE MOTION SEARCH RANGE DETERMINING APPARATUS AND METHOD 有权
    自适应运动搜索范围确定装置和方法

    公开(公告)号:US20120128261A1

    公开(公告)日:2012-05-24

    申请号:US13301093

    申请日:2011-11-21

    IPC分类号: G06K9/36

    CPC分类号: H04N19/57

    摘要: Provided is an adaptive motion search range determining apparatus and method for encoding UHD-class high-resolution images. The adaptive motion search range determining apparatus includes an MVD average/standard deviation calculation unit calculating an value average and a standard deviation of MVDs of neighboring macroblocks of a current macroblock, and a motion search range determination unit determining a motion search range of the current macroblock using the value average and the standard deviation. According to the adaptive motion search range determining apparatus, it is possible to enable each macroblock to have an adaptive search range by variably adjusting a motion vector search range of a current macroblock with reference to motion vectors of neighboring macroblocks of the current macroblock.

    摘要翻译: 提供了一种用于对UHD级高分辨率图像进行编码的自适应运动搜索范围确定装置和方法。 自适应运动搜索范围确定装置包括:MVD平均/标准偏差计算单元,用于计算当前宏块的相邻宏块的MVD的值平均值和标准偏差;以及运动搜索范围确定单元,确定当前宏块的运动搜索范围 使用值平均值和标准偏差。 根据自适应运动搜索范围确定装置,可以参考当前宏块的相邻宏块的运动矢量,通过可变地调整当前宏块的运动矢量搜索范围,使每个宏块具有自适应搜索范围。

    METHOD FOR EXTRACTING PROBABILITY MODEL VALUE FROM PROBABILITY MODEL TABLE AND METHOD AND APPARATUS FOR DECODING SYMBOL VALUE BY USING THE SAME
    6.
    发明申请
    METHOD FOR EXTRACTING PROBABILITY MODEL VALUE FROM PROBABILITY MODEL TABLE AND METHOD AND APPARATUS FOR DECODING SYMBOL VALUE BY USING THE SAME 审中-公开
    从可行性模型表提取概率模型值的方法和使用相同符号来解码符号值的方法和装置

    公开(公告)号:US20110153334A1

    公开(公告)日:2011-06-23

    申请号:US12969616

    申请日:2010-12-16

    IPC分类号: G10L19/00

    摘要: A method for extracting a probability model value from a probability model table and a method and apparatus for decoding a symbol value using the same are provided. The method for extracting a probability model value from a probability model table includes: segmenting and reducing a probability model table including a plurality of probability model values; disposing indexes on the basis of the segmented and reduced probability model table; and searching the probability model table for a probability model value by using the disposed indexes.

    摘要翻译: 提供了一种从概率模型表提取概率模型值的方法,以及用于使用该概率模型表解码符号值的方法和装置。 从概率模型表提取概率模型值的方法包括:分割并减少包括多个概率模型值的概率模型表; 在分段和简化概率模型表的基础上处理指标; 并通过使用所设置的索引来搜索概率模型表中的概率模型值。

    VIDEO DECODING APPARATUS AND METHOD BASED ON MULTIPROCESSOR
    7.
    发明申请
    VIDEO DECODING APPARATUS AND METHOD BASED ON MULTIPROCESSOR 审中-公开
    基于多处理器的视频解码设备和方法

    公开(公告)号:US20110085601A1

    公开(公告)日:2011-04-14

    申请号:US12836979

    申请日:2010-07-15

    IPC分类号: H04N7/26

    CPC分类号: H04N19/436 H04N19/44

    摘要: Disclosed are a multiprocessor-based video decoding apparatus and method. The multiprocessor-based video decoding apparatus includes: a stream parser dividing an input stream by row and parsing a skip counter and a quantization parameter of the input stream; and a plurality of processors acquiring the plurality of divided streams, the skip counter, and the quantization parameter generated by the stream parser, acquiring decoded information of an upper processor among neighboring processors by row, and parallel-decoding the plurality of divided streams by row. Decoding of an input stream can be parallel-processed by row.

    摘要翻译: 公开了一种基于多处理器的视频解码装置和方法。 基于多处理器的视频解码装置包括:流分析器,其逐行分割输入流,解析跳过计数器和输入流的量化参数; 以及多个处理器,获取多个划分的流,跳过计数器和由流解析器生成的量化参数,通过行获取相邻处理器之间的上位处理器的解码信息,并且按行并行解码多个划分的流 。 输入流的解码可以由行并行处理。

    VIDEO ENCODING APPARATUS AND METHOD USING PIPELINE TECHNIQUE WITH VARIABLE TIME SLOT
    10.
    发明申请
    VIDEO ENCODING APPARATUS AND METHOD USING PIPELINE TECHNIQUE WITH VARIABLE TIME SLOT 审中-公开
    视频编码设备和使用可变时隙的管道技术的方法

    公开(公告)号:US20090103625A1

    公开(公告)日:2009-04-23

    申请号:US12252734

    申请日:2008-10-16

    IPC分类号: H04N7/26

    CPC分类号: H04N19/436

    摘要: A video encoding apparatus and method using a pipeline technique with a variable time slot are provided. More particularly, a video encoding apparatus and method capable of shortening a video encoding time by variably adjusting lengths of time slots when an H.264 video encoding process is performed in a pipeline structure are provided. The video encoding apparatus includes a plurality of functional blocks that perform video encoding steps based on an H.264 standard for macroblocks configuring input digital video signals in a pipeline structure, and a controller that controls lengths of time slots configuring the pipeline structure based on done signals received from the plurality of functional blocks. Lengths of time slots can be adjusted according to operation times of video encoding steps using done signals generated from functional blocks, thereby preventing unnecessary power consumption and delays when using a fixed-length time slot.

    摘要翻译: 提供了使用具有可变时隙的流水线技术的视频编码装置和方法。 更具体地说,提供了一种视频编码装置和方法,其能够在流水线结构中执行H.264视频编码处理时,通过可变地调整时隙的长度来缩短视频编码时间。 视频编码装置包括:多个功能块,其基于用于在流水线结构中配置输入数字视频信号的宏块的H.264标准执行视频编码步骤;以及控制器,其基于完成来控制配置流水线结构的时隙的长度 从多个功能块接收的信号。 可以根据从功能块产生的完成信号的视频编码步骤的操作时间来调整时隙长度,从而在使用固定长度的时隙时防止不必要的功耗和延迟。