Video coder and corresponding method
This application relates to the field of video coding technologies, and discloses a video coder and a corresponding method, to help improve video coding performance. In this application, encoding and decoding are collectively referred to as coding. A video coding method includes: determining a block split policy of a current picture block based on a size relationship between the width and the height of the current picture block; applying the block split policy to the current picture block to obtain a coding block; and reconstructing the obtained coding block to reconstruct the current picture block.
1. A video coding method, comprising:
determining a block split policy of a current picture block based on a size relationship between a width and a height of the current picture block;
applying the block split policy to the current picture block to obtain a coding block; and
reconstructing the obtained coding block to reconstruct the current picture block;
wherein the determining of a block split policy of a current picture block based on a size relationship between a width and a height of the current picture block comprises:
determining whether the current picture block satisfies a first condition, wherein the first condition comprises: the width of the current picture block is less than a product of a first threshold and the height of the current picture block; or
determining whether the current picture block satisfies a second condition, wherein the second condition comprises: the height of the current picture block is less than a product of a first threshold and the width of the current picture block.
2. The method according to claim 1 , wherein
when the current picture block does not satisfy the first condition, determining that the block split policy is a split whose split direction is a vertical direction, wherein the vertical direction is perpendicular to a direction of a side on which the width of the current picture block is located.
3. The method according to claim 1 , wherein
when the current picture block does not satisfy the second condition, determining that the block split policy is a split whose split direction is a horizontal direction, wherein the horizontal direction is perpendicular to a direction of a side on which the height of the current picture block is located.
4. The method according to claim 1 , wherein
when the current picture block does not satisfy the first condition, determining that the block split policy of the current picture block does not comprise a split whose split direction is a horizontal direction, wherein the horizontal direction is perpendicular to a direction of a side on which the height of the current picture block is located.
5. The method according to claim 1 , wherein
when the current picture block does not satisfy the second condition, determining that the block split policy of the current picture block does not comprise a split whose split direction is a vertical direction, wherein the vertical direction is perpendicular to a direction of a side on which the width of the current picture block is located.
6. The method according to claim 2 , wherein the first threshold is an allowed maximum value of a ratio of a long-side length to a short-side length that are of a node in a coding tree.
7. The method according to claim 2 , further comprising:
parsing a bitstream to obtain identifier information, wherein the identifier information is used to indicate a split mode for splitting the current picture block; and
correspondingly, the applying of the block split policy to the current picture block to obtain a coding block comprises:
performing, on the current picture block according to the block split policy by using the split mode indicated by the identifier information, split whose split direction is the vertical direction, to obtain the coding block.
8. A video coder, comprising:
a split unit configured to determine a block split policy of a current picture block based on a size relationship between a width and a height of the current picture block, and apply the block split policy to the current picture block to obtain a coding block; and
a reconstruction unit configured to reconstruct the obtained coding block to reconstruct the current picture block, wherein the split unit is further configured to:
determine whether the current picture block satisfies a first condition, wherein the first condition comprises: the width of the current picture block is less than a product of a first threshold and the height of the current picture block; or
determine whether the current picture block satisfies a second condition, wherein the second condition comprises: the height of the current picture block is less than a product of a first threshold and the width of the current picture block.
9. The video coder according to claim 8 , wherein when the current picture block does not satisfy the first condition, the split unit is configured to determine that the block split policy is a split whose split direction is a vertical direction, wherein the vertical direction is perpendicular to a direction of a side on which the width of the current picture block is located.
10. The video coder according to claim 8 , wherein when the current picture block does not satisfy the second condition, the split unit is configured to determine that the block split policy is a split whose split direction is a horizontal direction, wherein the horizontal direction is perpendicular to a direction of a side on which the height of the current picture block is located.
11. The video coder according to claim 8 , wherein when the current picture block does not satisfy the first condition, the split unit is configured to determine that the block split policy of the current picture block does not comprise a split whose split direction is a horizontal direction, wherein the horizontal direction is perpendicular to a direction of a side on which the height of the current picture block is located.
12. The video coder according to claim 8 , wherein when the current picture block does not satisfy the second condition, the split unit is configured to determine that the block split policy of the current picture block does not comprise a split whose split direction is a vertical direction, wherein the vertical direction is perpendicular to a direction of a side on which the width of the current picture block is located.
13. The video coder according to claim 9 , wherein the first threshold is an allowed maximum value of a ratio of a long-side length to a short-side length that are of a node in a coding tree.
14. The video coder according to claim 9 , further comprising:
an entropy decoding unit, configured to parse a bitstream to obtain identifier information, wherein the identifier information is used to indicate a split mode for splitting the current picture block; and
correspondingly, the split unit is configured to perform, on the current picture block according to the block split policy by using the split mode indicated by the identifier information, split whose split direction is the vertical direction, to obtain the coding block.