Method for decoding image and apparatus using same
A method or coding image information, according to the present invention, comprises the steps of: binarizing according to different techniques, index values of forward prediction, backward prediction, and bidirectional prediction, depending on whether the bidirectional prediction is applied when inter-predicting a current block; and entropy coding a binarized codeword, wherein whether to apply the bidirectional prediction when inter-predicting the current block can be determined on the basis of the size of the current block. As a result, provided are a method for binarizing an inter-prediction direction of a prediction unit having a specific size, and an apparatus using same.
1. A method of decoding video, the method comprising:
receiving, by a decoding apparatus, a bitstream including inter prediction type information indicating whether list0 prediction, list1 prediction, or bi-prediction is applied to a current block;
deriving, by the decoding apparatus, a value of the inter prediction type information based on the bitstream;
determining, by the decoding apparatus, an inter prediction type for the current block based on the value of the inter prediction type information; and
performing, by the decoding apparatus, an inter prediction based on the inter prediction type,
wherein whether the bi-prediction is not available for the current block is determined based on a size of the current block.
2. The method of claim 1 , wherein when a current coding unit has an 8×8 size and a partition mode of the current block is 2N×N or N×2N, the bi-prediction is not available for the current block.
3. The method of claim 1 , wherein when the current block has a sum of width and length of 12, the bi-prediction is not available for the current block.
4. The method of claim 2 , wherein the current block is a prediction block.
5. The method of claim 1 , wherein if the bi-prediction is available for the current block, a bin string for the list0 prediction or the list1 prediction consists of 2 bits, and
wherein if the bi-prediction is not available for the current block, the bin string for the list0 prediction or the list1 prediction consists of 1 bit.
6. The method of claim 5 , wherein if the bi-prediction is available for the current block, the bin string for the bi-prediction consists of 1 bit.
7. An apparatus for decoding video, the apparatus comprising:
an entropy decoding module configured to receive a bitstream including inter prediction type information indicating whether list0 prediction, list1 prediction, or bi-prediction is applied to the current block, derive a value of the inter prediction type information based on the bitstream; and
a prediction module configured to determine an inter prediction type for the current block based on the value of the inter prediction type information, and preform an inter prediction based on the inter prediction type,
wherein whether the bi-prediction is not available for the current block is determined based on a size of the current block.
8. The apparatus of claim 7 , wherein when a current coding unit has an 8×8 size and a partition mode of the current block is 2N×N or N×2N, the bi-prediction is not available for the current block.
9. The apparatus of claim 7 , wherein when the current block has a sum of width and length of 12, the prediction is not available for the current block.
10. The apparatus of 8 , wherein the current block is a prediction block.
11. The apparatus of claim 7 , wherein if the bi-prediction is available for the current block, a bin string for the list0 prediction or the list1 prediction consists of 2 bits, and
wherein if the bi-prediction is not available for the current block, the bin string for the list0 prediction or the list1 prediction consists of 1 bit.
12. The apparatus of claim 11 , wherein if the bi-prediction is available for the current block, the bin string for the bi-prediction consists of 1 bit.
13. The method of claim 1 , wherein the list0 prediction is based on a reference picture list0, and the list1 prediction is based on a reference picture list1, and the bi-prediction is based on both the reference picture list0 and the reference picture list1.
14. The apparatus of claim 7 , wherein the list0 prediction is based on a reference picture list0, and the list1 prediction is based on a reference picture list1, and the bi-prediction is based on both the reference picture list0 and the reference picture list1.