| 摘要 |
Provided is a method in which an input image signal is divided into pixel blocks, and inter-prediction is carried out on the pixel blocks obtained by said division. This method includes selection of prediction motion information from within a motion information buffer that stores motion information of an encoded region, and use of the prediction motion information to predict the motion information of a block to be encoded. Furthermore, this method includes acquisition of representative motion information only, said representative motion information being acquired from among a plurality of motion information items within a region for which encoding has finished, in accordance with first information that indicates the selection method for the prediction motion information. |
