公开(公告)号：US20140104383A1

公开(公告)日：2014-04-17

申请号：US14125451

申请日：2012-06-14

申请人： Takahashi Yoshitomo ,  Shinobu Hattori

发明人： Takahashi Yoshitomo ,  Shinobu Hattori

IPC分类号： H04N13/00 ,  H04N19/51 ,  H04N19/597

CPC分类号： H04N13/161 ,  H04N19/463 ,  H04N19/52 ,  H04N19/597

摘要： The present technique relates to an image processing device and method that can increase encoding efficiency in multi-view encoding. A predicted vector generation unit generates a predicted vector by using a motion disparity vector of a peripheral region located in the vicinity of the current region. When a predicted vector of a disparity vector is to be determined, but it is not possible to refer to any of the peripheral regions at this point, the predicted vector generation unit sets the minimum disparity value or the maximum disparity value supplied from a disparity detection unit as the predicted vector. The present disclosure can be applied to image processing devices, for example.

摘要翻译： 本技术涉及可以提高多视图编码中的编码效率的图像处理装置和方法。 预测矢量生成单元通过使用位于当前区域附近的周边区域的运动视差矢量来生成预测矢量。 当要确定视差向量的预测向量时，在这一点上不可能参考任何外围区域，则预测向量生成单元设置从视差检测提供的最小视差值或最大视差值 单位作为预测向量。 本公开可以应用于例如图像处理装置。