首页> 外国专利> MOTION COMPENSATION MOVING IMAGE CODER AND MOTION COMPENSATION MOVING IMAGE DECODER

MOTION COMPENSATION MOVING IMAGE CODER AND MOTION COMPENSATION MOVING IMAGE DECODER

机译:运动补偿运动图像编码器和运动补偿运动图像解码器

摘要

PPROBLEM TO BE SOLVED: To provide a method for motion estimation with adaptive motion accuracy including technique for computing motion vectors of high pixel accuracy with a minor increase in computation. PSOLUTION: A coder carries out motion compensation by using the motion vectors expressed by a plurality of fractional accuracies expressed by 1/N pel (N is an integer), performs variable-length coding for accuracy apart from the motion vectors and codes the motion vectors at everty block. Motion compensation is carried out by using a second filter of more computational complexity according to second accuracy of higher accuracy after execution according to first accuracy by a first filter. In a decoder, coded data is decoded. Accuracy is set at frame unit. Although the same accuracy is used for all the motion vectors in one frame, different accuracies are used for different frames. A motion vector of a coding object is equivalent to a product of accuracy to vector values of coded data and is interpreted as different values according to accuracy. PCOPYRIGHT: (C)2011,JPO&INPIT
机译:

要解决的问题:提供一种具有自适应运动精度的运动估计方法,包括用于以较小的计算量来计算高像素精度的运动矢量的技术。

解决方案:编码器通过使用由1 / N pel(N为整数)表示的多个小数精度表示的运动矢量执行运动补偿,除运动矢量和代码外还执行精度可变的长度编码运动矢量在every block。在通过第一滤波器根据第一精度执行之后,通过使用根据更高精度的第二精度的,具有更高计算复杂度的第二滤波器来执行运动补偿。在解码器中,对编码数据进行解码。精度以帧为单位设置。尽管对于一帧中的所有运动矢量使用相同的精度,但是对不同的帧使用不同的精度。编码对象的运动矢量等效于精度与编码数据的矢量值的乘积,并且根据精度被解释为不同的值。

版权:(C)2011,日本特许厅&INPIT

著录项

  • 公开/公告号JP2011035928A

    专利类型

  • 公开/公告日2011-02-17

    原文格式PDF

  • 申请/专利权人 SHARP CORP;

    申请/专利号JP20100217117

  • 发明设计人 RIBAS JORDI;SHEN JIANDONG;

    申请日2010-09-28

  • 分类号H04N7/32;

  • 国家 JP

  • 入库时间 2022-08-21 18:24:16

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号