In one embodiment, the method includes obtaining reference picture information, and determining a first reference picture of the bi-predictive image block. The first reference picture has a co-located image block with respect to the bi-predictive image block. A second reference picture is determined based on a motion vector of the co-located image block. At least one motion vector of the bi-predictive image block is obtained according to the motion vector of the co-located image block and the reference picture type of at least one of the first reference picture and the second reference picture. The reference picture information indicates reference picture types of reference pictures, and the reference picture type being based on temporal distance between a current picture including the bi-predictive image block and the reference picture. The bi-predictive image block is decoded by motion-estimating the bi-predictive image block using the motion vector of the bi-predictive image block.