# US 11,445,213 B2 · Inter-prediction-based image or video coding using sbTMVP > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,445,213 B2 - Document ID: US11445213B2 - Status: Granted - Kind: B2 - Application number: 17/545,927 - Filed: 2021-12-08 - Published: 2022-03-31 - Granted: 2022-09-13 - Earliest claimed: 2019-06-13 - Assignees: LG Electronics Inc. - Applicants: LG Electronics Inc. - Inventors: Hyeongmoon Jang; Naeri Park; Junghak Nam - Web page: https://patentgrep.com/patent/US11445213B2 ## CPC - H04N 19/52: by predictive encoding - H04N 19/105: Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction - 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 19/159: Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction - H04N 19/167: Position within a video image, e.g. region of interest [ROI] - H04N 19/176: the region being a block, e.g. a macroblock - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract According to the disclosure of the present document, in a subblock-based temporal motion vector prediction (sbTMVP), a position of a subblock for deriving a motion vector per subblock unit can be efficiently calculated, and thus video/image coding efficiency can be increased and hardware implementation can be simplified. ## Claims ### Claim 1 An image decoding method performed by a decoding apparatus, the method comprising: deriving a position of a current subblock within a current block; deriving a reference subblock on a collocated reference picture based on the position of the current subblock; deriving a subblock temporal motion vector predictor (sbTMVP) candidate based on a motion vector of the reference subblock; deriving a motion vector of the current subblock based on the sbTMVP candidate; generating prediction samples of the current block based on the motion vector of the current subblock; and generating reconstruction samples based on the prediction samples, wherein the position of the current subblock is derived based on a center sample position of the current subblock, wherein the reference subblock is derived on the collocated reference picture based on the center sample position of the current subblock, wherein the center sample position represents a position of a right-lower center sample positioned at a bottom right, among four samples positioned at the center, wherein the sbTMVP candidate is derived based on the motion vector of the reference subblock derived based on an availability of the reference subblock, wherein for the reference subblock that is available, a motion vector of the available reference subblock is derived as the sbTMVP candidate, wherein the motion vector of the available reference subblock is derived based on a motion vector of a block covering a modified location derived based on a top-left sample position of the reference subblock, and wherein the modified location is derived by an equation ((xColSb>>3)<<3, (yColSb>>3)<<3), where the xColSb and the yColSb represent the x coordinate and y coordinate of the sample position of the reference subblock derived based on the center sample position of the current subblock, respectively, and >> represents an arithmetic right shift, and << represents an arithmetic left shift. ### Claim 2 (depends on 1) The image decoding method of claim 1, wherein for the reference subblock that is not available, a base motion vector is derived as the sbTMVP candidate. ### Claim 3 (depends on 2) The image decoding method of claim 2, wherein the base motion vector is derived on the collocated reference picture based on the position of the current block, wherein the position of the current block is derived based on a center sample position of the current block. ### Claim 4 (depends on 2) The image decoding method of claim 2, wherein the unavailable reference subblock includes a reference subblock positioned outside the collocated reference picture or a reference subblock whose motion vector is not available. ### Claim 5 (depends on 4) The image decoding method of claim 4, wherein, for the reference subblock that is an intra mode, an IBC (intra block copy) mode, or a palette mode, the reference subblock is a subblock whose the motion vector is not available. ### Claim 6 (depends on 1) The image decoding method of claim 1, wherein the deriving of the reference subblock comprises: performing a motion shift on the center sample position of the current subblock based on a motion vector derived from a left neighboring block of the current block; and deriving the reference subblock on the collocated reference picture based on the motion-shifted position. ### Claim 7 An image encoding method performed by an encoding apparatus, the method comprising: deriving a position of a current subblock within a current block; deriving a reference subblock on a collocated reference picture based on the position of the current subblock; deriving a subblock temporal motion vector predictor (sbTMVP) candidate based on a motion vector of the reference subblock; deriving a motion vector of the current subblock based on the sbTMVP candidate; generating prediction samples of the current block based on the motion vector of the current subblock; generating information on residual samples derived based on the prediction samples; and encoding image information including the information on the residual samples, wherein the position of the current subblock is derived based on a center sample position of the current subblock, wherein the reference subblock is derived on the collocated reference picture based on the center sample position of the current subblock, wherein the center sample position represents a position of a right-lower center sample positioned at a bottom right, among four samples positioned at the center, wherein the sbTMVP candidate is derived based on the motion vector of the reference subblock derived based on the availability of the reference subblock, wherein for the reference subblock that is available, a motion vector of the available reference subblock is derived as the sbTMVP candidate, wherein the motion vector of the available reference subblock is derived based on a motion vector of a block covering a modified location derived based on a top-left sample position of the reference subblock, and wherein the modified location is derived by an equation ((xColSb>>3)<<3, (yColSb>>3)<<3), where the xColSb and the yColSb represent the x coordinate and y coordinate of the sample position of the reference subblock derived based on the center sample position of the current subblock, respectively, and >> represents an arithmetic right shift, and << represents an arithmetic left shift. ### Claim 8 (depends on 7) The image encoding method of claim 7, wherein for the reference subblock that is not available, a base motion vector is derived as the sbTMVP candidate. ### Claim 9 (depends on 8) The image encoding method of claim 8, wherein the base motion vector is derived on the collocated reference picture based on the position of the current block, wherein the position of the current block is derived based on a center sample position of the current block. ### Claim 10 (depends on 8) The image encoding method of claim 8, wherein the unavailable reference subblock includes a reference subblock positioned outside the collocated reference picture or a reference subblock whose motion vector is not available. ### Claim 11 (depends on 10) The image encoding method of claim 10, wherein, for the reference subblock that is an intra mode, an IBC (intra block copy) mode, or a palette mode, the reference subblock is a subblock whose the motion vector is not available. ### Claim 12 (depends on 7) The image encoding method of claim 7, wherein the deriving of the reference subblock comprises: performing a motion shift on the center sample position of the current subblock based on a motion vector derived from a left neighboring block of the current block; and deriving the reference subblock on the collocated reference picture based on the motion-shifted position. ### Claim 13 A non-transitory computer-readable digital storage medium storing a bitstream generated by a method, the method comprising: deriving a position of a current subblock within a current block; deriving a reference subblock on a collocated reference picture based on the position of the current subblock; deriving a subblock temporal motion vector predictor (sbTMVP) candidate based on a motion vector of the reference subblock; deriving a motion vector of the current subblock based on the sbTMVP candidate; generating prediction samples of the current block based on the motion vector of the current subblock; generating information on residual samples derived based on the prediction samples; and encoding image information including the information on the residual samples to generate the bitstream, wherein the position of the current subblock is derived based on a center sample position of the current subblock, wherein the reference subblock is derived on the collocated reference picture based on the center sample position of the current subblock, wherein the center sample position represents a position of a right-lower center sample positioned at a bottom right, among four samples positioned at the center, wherein the sbTMVP candidate is derived based on the motion vector of the reference subblock derived based on the availability of the reference subblock, wherein for the reference subblock that is available, a motion vector of the available reference subblock is derived as the sbTMVP candidate, wherein the motion vector of the available reference subblock is derived based on a motion vector of a block covering a modified location derived based on a top-left sample position of the reference subblock, and wherein the modified location is derived by an equation ((xColSb>>3)<<3, (yColSb>>3)<<3), where the xColSb and the yColSb represent the x coordinate and y coordinate of the sample position of the reference subblock derived based on the center sample position of the current subblock, respectively, and >> represents an arithmetic right shift, and << represents an arithmetic left shift.