Method for inducing a merge candidate block and device using same

A method of decoding a video signal, comprising:

• determining whether a spatial merge candidate is decoded or not;
• determining, based on parallel merge level information, whether the spatial merge candidate is included in a same MER (Motion Estimation Region) as a current prediction block;
• when the spatial merge candidate is decoded and is not included in the same MER as the current prediction block, determining that the spatial merge candidate is an available merge candidate for inter prediction of the current prediction block;
• deriving motion information from the available merge candidate block, the motion information comprising at least one of motion vector information or reference picture information;
• generating a merge candidate list of the current prediction block, the merge candidate list comprising the available merge candidate;
• and
• performing an inter prediction relating to the current block based on the merge candidate list,
• wherein the spatial merge candidate block includes at least one of neighboring blocks adjacent to the current prediction block, the neighboring blocks including a left neighboring block, a top neighboring block, a top-right neighboring block, a left-bottom neighboring block or a top-left neighboring block,
• wherein when a size of a MER (Motion Estimation Region) is 8×8 and a size of a coding block is 8×8, at least one of the neighboring blocks is replaced to a replacement block located outside the MER, and
• wherein when the size of the MER is not 8×8 and the size of the coding block is not 8×8, none of the neighboring blocks is replaced to the replacement block located outside the MER.