发明名称 Inter prediction method and apparatus using adjacent pixels, and image encoding/decoding method and apparatus using same
摘要 The present disclosure relates to an inter prediction method and apparatus using adjacent pixels, and a video encoding method and apparatus using the same. The inter prediction method for a predictive encoding includes: determining a current motion vector by estimating a motion of a current block; generating a reference block indicated by the current motion vector; calculating a motion compensation coefficient or an average compensation coefficient by using pixels neighboring the current block and pixels neighboring reference block; generating a changed reference block reflecting the motion compensation coefficient or the average compensation coefficient in the reference block; and determining the changed reference block as a predicted block of the current block. The present disclosure reduces a difference between an actual block and a predicted block by more accurately predicting the block to be encoded, thereby improving the compression efficiency.
申请公布号 US9571851(B2) 申请公布日期 2017.02.14
申请号 US201013497928 申请日期 2010.09.24
申请人 SK TELECOM CO., LTD. 发明人 Kim Sunyeon;Lim Jeongyeon;Lee Yunglyul;Moon Joohee;Kim Haekwang;Jeon Byeungwoo;Han Jongki;Seo Jeonghoon;Kim Daeyeon;Hong Sungwook;Choi Jaehoon;Lee Gyumin
分类号 H04N7/12;H04N11/02;H04N11/04;H04N19/51;H04N19/105;H04N19/176;H04N19/147 主分类号 H04N7/12
代理机构 Hauptman Ham, LLP 代理人 Hauptman Ham, LLP
主权项 1. An inter prediction method for encoding a video image, comprising: determining a current motion vector of a current block in a current picture; generating a reference block by interpolating pixels at a position indicated by the current motion vector in a reference picture; calculating a compensation coefficient by using values of one or more first pixels neighboring the current block in the current picture and values of one or more second pixels neighboring the reference block in the reference picture; when a value of the compensation coefficient is not “0”, generating a changed reference block by applying the compensation coefficient to the reference block, andsetting the changed reference block to be a predicted block of the current block; and when the value of the compensation coefficient is “0”, setting the reference block to be the predicted block of the current block, wherein the one or more first pixels and the one or more second pixels are pixels encoded prior to encoding the current block.
地址 Seoul KR