| 摘要 |
A method for determining a motion vector is provided to improve overall coding efficiency by differently determining the direct mode forward and backward motion vectors of a B(Bi-predictive) picture according to a reference buffer that a motion vector, produced from a specific picture, indicates and making direct mode frequently used as coding mode in coding B pictures. In decoding a bi-predictive image block, a decoder acquires reference picture information. In this case, the reference picture information indicates a reference picture type which is based on the temporal distance between the current picture and a reference picture. The decoder determines the first reference picture, which has a co-located image block for the bi-predictive image block, for the bi-predictive image block. Based on motion vectors of the co-located image block, the decoder determines the second reference picture. The decoder acquires one or more motion vectors of the bi-predictive image according to motion vectors of the co-located image block and the reference picture type of either the first reference picture or the second reference picture. The decoder decodes the bi-predictive image block by judging the movement of the bi-predictive image block using the acquired motion vectors.
|
