# US 11,445,186 B2 · Method and apparatus for processing video signal > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,445,186 B2 - Document ID: US11445186B2 - Status: Granted - Kind: B2 - Application number: 16/463,823 - Filed: 2017-11-23 - Published: 2019-12-19 - Granted: 2022-09-13 - Earliest claimed: 2016-11-25 - Assignees: Kt Corporation - Applicants: Kt Corporation - Inventors: Bae Keun Lee - Web page: https://patentgrep.com/patent/US11445186B2 ## CPC - H04N 19/119: Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks - H04N 19/176: the region being a block, e.g. a macroblock - H04N 19/46: Embedding additional information in the video signal during the compression process (H04N19/517, H04N19/68, H04N19/70 take precedence) - H04N 19/70: characterised by syntax aspects related to video coding, e.g. related to compression standards - H04N 19/91: Entropy coding, e.g. variable length coding [VLC] or arithmetic coding - H04N 19/96: Tree coding, e.g. quad-tree coding - H04N 19/109: among a plurality of temporal predictive coding modes - H04N 19/11: among a plurality of spatial predictive coding modes - H04N 19/132: Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A method for decoding a video according to the present invention may comprise: decoding partition information indicating whether a current coding block is to be divided by a partitioning line in a vertical direction or a partitioning line in a horizontal direction, and dividing the coding block into at least one sub-block based on the partition information. ## Claims ### Claim 1 A method for decoding a video, the method comprising: determining whether a coding block is partitioned into two transform blocks of rectangle shapes; when it is determined to partition the coding block, determining, based on a first flag of 1-bit decoded from a bitstream, whether partitioning the coding block into the two transform blocks is performed symmetrically or asymmetrically; determining, based on a second flag of 1-bit decoded from the bitstream, whether partitioning the coding block into the two transform blocks is performed in a horizontal direction or in a vertical direction; partitioning the coding block into a first transform block and a second transform block; determining a transform type for the first transform block; and performing an inverse-transform for the first transform block based on the transform type, wherein when it is determined to partition the coding block asymmetrically, the coding block is partitioned into the first transform block of ¼ size of the coding block and the second transform block of ¾ size of the coding block, wherein locations of the first transform block and the second transform block are determined based on a third flag of 1-bit decoded from the bitstream, wherein when it is determined to partition the coding block in the horizontal direction, it is determined based on the third flag which of the first transform block and the second transform block is located upper side of the coding block, wherein when it is determined to partition the coding block in the vertical direction, it is determined based on the third flag which of the first transform block and the second transform block is located right side of the coding block, wherein whether to decode a transform skip flag, indicating whether the inverse-transform is skipped or not, for the first transform block from the bitstream is determined based on whether the coding block is partitioned into the two transform blocks or not, and wherein in response to the coding block being partitioned into the two transform blocks, decoding the transform skip flag for the first transform block from the bitstream is omitted and the inverse transform block is forced not to skip the inverse-transform for the first transform block irrelevant to a size and a shape of the first transform block. ### Claim 2 (depends on 1) The method of claim 1, wherein in response to the coding block being partitioned into the first transform block and the second transform block, first residual samples in the first transform block are derived by performing the inverse-transform on inverse-quantized residual coefficients for the first transform block, and second residual samples in the second transform block are derived to be a default value, and wherein the default value is representative of 0. ### Claim 3 A method for encoding a video, the method comprising: determining whether to partition a coding block into a first transform block and a second transform block; encoding a first flag of 1-bit indicating whether the coding block is partitioned into two transform blocks or not; encoding a second flag of 1-bit specifying whether partitioning the coding block into the two transform blocks is performed symmetrically or asymmetrically; encoding a third flag of 1-bit specifying whether partitioning the coding block into the two transform blocks is performed in a horizontal direction or in a vertical direction; determining a transform type for the first transform block; and performing a transform for the first transform block based on the transform type, wherein when partitioning the coding block is performed asymmetrically, the coding block is partitioned into the first transform block of ¼ size of the coding block and the second transform block of ¾ size of the coding block, wherein a fourth flag of 1-bit which is used to determine locations of the first transform block and the second transform block is encoded into a bitstream, wherein when the coding block is partitioned in the horizontal direction, the fourth flag is used to determine which of the first transform block and the second transform block is located upper side of the coding block, wherein when the coding block is partitioned in the vertical direction, the fourth flag is used to determine which of the first transform block and the second transform block is located right side of the coding block, wherein whether to encode a transform skip flag, indicating whether the transform is skipped or not, for the first transform block into the bitstream is determined based on whether the coding block is partitioned into the two transform blocks or not, and wherein in response to the coding block being partitioned into the two transform blocks, it is forced not to skip the transform for the first transform block irrelevant to a size and a shape of the first transform block, and encoding the transform skip flag for the first transform block into the bitstream is omitted. ### Claim 4 (depends on 3) The method of claim 3, wherein in response to the coding block being partitioned into the first transform block and the second transform block, first residual samples of the first transform block are encoded into the bitstream while second residual samples of the second transform block are not encoded into the bitstream and are set equal to a default value, and wherein the default value is representative of 0. ### Claim 5 A non-transitory computer readable medium for storing compressed video data associated with a video signal, comprising: a data stream stored in the non-transitory computer-readable medium, the data stream comprising: a first flag of 1-bit indicating whether a coding block is partitioned into two transform blocks or not; a second flag of 1-bit specifying whether partitioning the coding block into the two transform blocks is performed symmetrically or asymmetrically; and a third flag of 1-bit specifying whether partitioning the coding block into the two transform blocks is performed in a horizontal direction or in a vertical direction, wherein when the coding block is partitioned asymmetrically, the coding block is partitioned into a first transform block of ¼ size of the coding block and a second transform block of ¾ size of the coding block, wherein a transform type for the first transform block is determined and an inverse-transform for the first transform block is performed based on the transform type, wherein a fourth flag of 1-bit which is used to determine locations of the first transform block and the second transform block is further comprised in the data stream, wherein when the coding block is partitioned in the horizontal direction, the fourth flag is used to determine which of the first transform block and the second transform block is located upper side of the coding block, wherein when the coding block is partitioned in the vertical direction, the fourth flag is used to determine which of the first transform block and the second transform block is located right side of the coding block, wherein whether the data stream further comprises a transform skip flag, indicating whether the inverse-transform is skipped or not, for the first transform block is determined based on whether the coding block is partitioned into the two transform blocks or not, and wherein in response to the coding block being partitioned into the two transform blocks, the transform skip flag for the first transform block is not present in the data stream and the inverse transform block is forced not to skip the inverse-transform for the first transform block irrelevant to a size and a shape of the first transform block. ### Claim 6 (depends on 1) The method of claim 1, wherein under a case that the coding block has a first size, both symmetric partitioning and asymmetric partitioning are allowed for the coding block, and wherein under a case that the coding block has a second size different from the first size, only the symmetric partitioning is allowed for the coding block but the asymmetric partitioning is not allowed for the coding block. ### Claim 7 (depends on 1) The method of claim 1, wherein the method further comprises: determining whether geometric partitioning is applied to the coding block; partitioning the coding block into a first prediction unit and a second prediction unit in response to a determination that the geometric partitioning is applied to the coding block; and deriving prediction information for each of the first prediction unit and the second prediction unit, and wherein the prediction information for the second prediction unit is derived by referring to the prediction information for the first prediction unit.