Method for encoding and decoding image and device using same

A decoding apparatus, comprising:

• a predictor configured to determine whether a first pixel belongs to a Largest Coding Unit (LCU) to which a co-located block belongs, wherein the first pixel neighbors the co-located block and located at a lower right side of the co-located block, wherein the co-located block corresponds to a current block and is located in a reference picture, to determine a modified location in the reference picture, determining a reference prediction unit (colPu), which is a prediction block encompassing the modified location, to derive a temporal motion vector predictor (TMVP) for the current block from motion information of the determined reference prediction unit, to derive a motion vector of the current block based on the TMVP, to generate a predicted pixel of the current block based on the motion vector, and to generate a reconstructed picture based on the predicted pixel;
• and
• a memory configured to store the reference picture,
• wherein based on the first pixel not belonging to the LCU, the modified location is specified by a position ((xPCtr>>4)<<4, (yPCtr>>4)<<4), and wherein a position (xPCtr, yPCtr) is a position of a second pixel located at the bottom right side among four central pixels of the co-located block.