Method for generating prediction block in AMVP mode

A non-transitory recording medium which includes a bit stream decoded by an image decoding device, the bit stream being generated by dividing an image into a plurality of blocks and by performing intra prediction or inter prediction for the block and including:

• information indicating at least an intra mode index, motion information, a quantization parameter and a quantized coefficient sequence,
• wherein the image decoding device,
• separates the at least an intra mode index, the motion information, the quantization parameter and the quantized coefficient sequence from the coded bit stream by decoding the coded bit stream,
• converts the quantized coefficient sequence to a quantization block, inversely quantizes coefficients of the quantization block using the quantization parameter to generate a transform block, and inversely transforms the transform block to generate a residual block,
• restores an intra prediction mode of a current prediction unit using the intra mode index and generates a prediction block of the current prediction unit according to the intra prediction mode when the coded bit stream is coded in the intra prediction mode, and
• generates a prediction block of a current prediction unit using a reference index and a motion vector derived from the motion information when the coded bit stream is coded in an inter prediction mode,
• wherein the motion vector is decoded using a motion vector predictor which is set equal to one of an effective spatial merge candidate and a temporal merge candidate,
• wherein a motion vector of the temporal merge candidate is a motion vector of a temporal merge candidate within a temporal merge candidate picture, and the quantization parameter is encoded using an average of two effective quantization parameters among a left quantization parameter, an upper quantization parameter and a previous quantization parameter of a current coding unit,
• wherein when the quantized block is larger than a predetermined size, the quantized coefficient sequence is inversely scanned in a unit of sub-block using a scan pattern and a plurality of sub-blocks are inversely scanned using a scan pattern, and the scan pattern for inversely scanning the plurality of sub-blocks is the same as the scan pattern for inversely scanning the quantized coefficients sequence, and
• wherein the scan pattern for inversely scanning the quantized coefficients is determined according to the intra prediction mode and a size of a transform unit when the coded bit stream is coded in the intra prediction mode.