| 摘要 |
Disclosed are a method of deciding candidate blocks in the prediction between images and an apparatus using the same. A method of decoding an image may include the steps of inserting a second block, which is located at a left lower end, into a merge candidate list when the second block at the left lower end is available; inserting a second block, which is located at a right upper end, into the merge candidate list when the second block at the right upper end is available; inserting a first block, which is located at the left lower end, into the merge candidate list when the first block at the left lower end is available; inserting a first block, which is located at the right upper end, into the merge candidate list when the first block at the right upper end is available; inserting a second block, which is located at a left upper end, into the merge candidate list when the second block at the left upper end is available; and inserting a temporal candidate prediction block into the merge candidate list when the temporal candidate prediction block is available. Accordingly, the encoding efficiency can be enhanced, and the complexity can be reduced. [Reference numerals] (AA) Start; (BB) End; (S500) Insert a left lower second block into a merge candidate block list in case the left lower second block is available; (S510) Insert a right upper second block into the merge candidate block list in case the right upper second block is available; (S520) Insert a left lower first block into the merge candidate block list in case the left lower first block is available; (S530) Insert a right upper first block into the merge candidate block list in case the right upper first block is available; (S540) Insert a left upper second block into the merge candidate block list in case the left upper second block is available; (S550) Insert a time candidate prediction block into the merge candidate list in case the time candidate prediction block is available |
