| CPC H04N 19/52 (2014.11) [H04N 19/577 (2014.11)] | 20 Claims | 

| 
               1. A video decoding method comprising the steps of: 
            generating a merge candidate list for a current block; 
                determining a merge candidate for the current block among merge candidates included in the merge candidate list; 
                deriving an offset vector for the current block, 
                wherein a magnitude of the offset vector is determined based on first index information specifying one among motion magnitude candidates, 
                wherein the magnitude of the offset vector is obtained by applying an arithmetic left shift operation by two binary digits to a value indicated by the motion magnitude candidate specified by the first index information; and 
                deriving a motion vector for the current block by adding the offset vector to a motion vector of the merge candidate; 
                wherein at least one among a maximum numerical value and a minimum numerical value of the motion magnitude candidates is set differently according to a numerical value of a flag indicating a numerical range of the motion magnitude candidates, and 
                the flag is signalled at a picture level. 
               |