In one embodiment, a method of predicting a motion vector for a current block in a current picture by a moving picture decoding device includes obtaining, by the moving picture decoding device, a direction of a reference picture of the current picture, based on comparison of a display order of the reference picture and a display order of the current picture, obtaining, by the moving picture decoding device, at least three forward motion vectors for at least three other blocks than the current block based on the direction of the reference picture, and predicting, by the moving picture decoding device, a forward motion vector for the current block by using median operation of the at least three forward motion vectors.