# US 11,909,953 B2 · Representation of affine model > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,909,953 B2 - Document ID: US11909953B2 - Status: Granted - Kind: B2 - Application number: 17/193,397 - Filed: 2021-03-05 - Published: 2021-07-08 - Granted: 2024-02-20 - Earliest claimed: 2018-09-23 - Assignees: Beijing Bytedance Network Technology Co., Ltd.; Bytedance Inc. - Applicants: Beijing Bytedance Network Technology Co., Ltd.; Bytedance Inc. - Inventors: Kai Zhang; Li Zhang; Hongbin Liu; Yue Wang - Web page: https://patentgrep.com/patent/US11909953B2 ## CPC - 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 19/537: Motion estimation other than block-based - H04N 19/137: Motion inside a coding unit, e.g. average field, frame or block difference - H04N 19/176: the region being a block, e.g. a macroblock - H04N 19/30: using hierarchical techniques, e.g. scalability (H04N19/63 takes precedence) - H04N 19/513: Processing of motion vectors - H04N 19/521: {for estimating the reliability of the determined motion vectors or motion vector field, e.g. for smoothing the motion vector field or for correcting motion vectors} - H04N 19/527: Global motion vector estimation - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A method, a video processing apparatus and computer-readable program medium for video process is disclosed. A method of video processing includes storing, during a conversion between a current block of video and a bitstream representation of the current block, affine model parameters for the current block; and performing the conversion using the affine model parameters, wherein the affine model parameters are derived at least from motion vectors of control points of the current block. ## Claims ### Claim 1 A method of processing video data, comprising: deriving, during a conversion between a current block of a video and a bitstream of the current block, at least one combined affine model using multiple original affine models with control point motion vectors, wherein each of the at least one combined affine model comprises a first set of control point motion vectors, wherein the multiple original affine models have multiple second sets of control point motion vectors, respectively, and wherein each control point motion vector in the first set of control point motion vectors is a function of corresponding control point motion vectors in each of the multiple second sets of control point motion vectors; storing affine model parameters for the current block; and performing the conversion based on the affine model parameters, wherein the affine model parameters are signaled from an encoder to a decoder, wherein the affine model parameters are derived at least from motion vectors of control points of the current block, and wherein the conversion is performed without storing the control point motion vectors of the current block. ### Claim 2 (depends on 1) The method of claim 1, wherein a number of stored affine model parameters is based on an affine model. ### Claim 3 (depends on 1) The method of claim 1, wherein the storing the affine model parameters includes storing two or four parameters for a 4-parameter affine model. ### Claim 4 (depends on 1) The method of claim 1, wherein the storing the affine model parameters includes storing four or six parameters for a 6-parameter affine model. ### Claim 5 (depends on 1) The method of claim 1, further comprising storing one of multiple control point motion vectors of the current block. ### Claim 6 (depends on 1) The method of claim 1, wherein the performing the conversion includes deriving affine motion for sub-blocks of the current block from the affine model parameters. ### Claim 7 (depends on 1) The method of claim 1, wherein the performing the conversion includes deriving affine motion for sub-blocks of the current block from the affine model parameters and one of multiple control point motion vectors. ### Claim 8 (depends on 7) The method of claim 7, wherein a number of the affine model parameters used to derive the affine motion for the sub-blocks of the current block is based on an affine model. ### Claim 9 (depends on 8) The method of claim 8, wherein the affine motion for the sub-blocks of the current block is derived as a normalization function based on one of: the affine model parameters, and the affine model parameters adding one of the multiple control point motion vectors. ### Claim 10 (depends on 1) The method of claim 1, wherein the affine model parameters for the current block are predicted from parameters of neighboring affine block. ### Claim 11 (depends on 1) The method of claim 1, wherein one control point motion vector of multiple control point motion vectors is signaled for the current block. ### Claim 12 (depends on 11) The method of claim 11, wherein the one control point motion vector is predicted by a motion vector prediction, and a Motion Vector Difference (MVD) is signaled as a normal motion vector. ### Claim 13 (depends on 1) The method of claim 1, wherein the affine model parameters for the current block are derived from a spatial or temporal neighboring block of the current block. ### Claim 14 (depends on 13) The method of claim 13, wherein one of multiple control point motion vectors is derived from the spatial or temporal neighboring block of the current block. ### Claim 15 (depends on 1) The method of claim 1, wherein the conversion comprises reconstructing the current block from the bitstream. ### Claim 16 (depends on 1) The method of claim 1, wherein the conversion comprises generating the bitstream from the current block. ### Claim 17 An apparatus in a video system comprising a processor and a non-transitory memory with instructions thereon, wherein the instructions upon execution by the processor, cause the processor to: derive, during a conversion between a current block of a video and a bitstream of the current block, at least one combined affine model using multiple original affine models with control point motion vectors, wherein each of the at least one combined affine model comprises a first set of control point motion vectors, wherein the multiple original affine models have multiple second sets of control point motion vectors, respectively, and wherein each control point motion vector in the first set of control point motion vectors is a function of corresponding control point motion vectors in each of the multiple second sets of control point motion vectors; store affine model parameters for the current block; and perform the conversion based on the affine model parameters, wherein the affine model parameters are signaled from an encoder to a decoder, wherein the affine model parameters are derived at least from motion vectors of control points of the current block, and wherein the conversion is performed without storing the control point motion vectors of the current block. ### Claim 18 A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises: deriving, for a current block of the video, at least one combined affine model using multiple original affine models with control point motion vectors, wherein each of the at least one combined affine model comprises a first set of control point motion vectors, wherein the multiple original affine models have multiple second sets of control point motion vectors, respectively, and wherein each control point motion vector in the first set of control point motion vectors is a function of corresponding control point motion vectors in each of the multiple second sets of control point motion vectors; storing affine model parameters for the current block; and generating the bitstream from the current block based on the affine model parameters, wherein the affine model parameters are signaled from an encoder to a decoder, wherein the affine model parameters are derived at least from motion vectors of control points of the current block, and wherein the generating is performed without storing control point motion vectors of the current block.