摘要 |
The present invention discloses a method for predicting a motion vector for coding a video or decoding a video, and an apparatus for same. Disclosed is a method comprising: determining a plurality of candidate blocks, which are referenced for predicting the motion vector of a current block from among neighboring blocks of the current block; determining a candidate motion vector of a first candidate block on the basis of whether each of a reference picture of the first candidate block from among the plurality of candidate blocks and a reference picture of the current block is a long-term reference picture; including the candidate motion vector of the first candidate block that is determined; and determining the motion vector of the current block by using a candidate motion vector list, which includes candidate motion vectors of remaining candidate blocks. |