# US 12,225,198 B2 · MIP for all channels in the case of 4:4:4-chroma format and of single tree > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,225,198 B2 - Document ID: US12225198B2 - Status: Granted - Kind: B2 - Application number: 17/916,574 - Filed: 2021-04-01 - Published: 2023-06-15 - Granted: 2025-02-11 - Earliest claimed: 2020-04-02 - Assignees: Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. - Applicants: Fraunhofer-Gesellschaft Zur Förderung Der Angewandten Forschung E.V. - Inventors: Jonathan Pfaff; Tobias Hinz; Philipp Helle; Philipp Merkle; Björn Stallenberger; Michael Schäfer; Benjamin Bross; Heiko Schwarz; Detlev Marpe; Thomas Wiegand - Web page: https://patentgrep.com/patent/US12225198B2 ## CPC - 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 19/593: involving spatial prediction techniques - 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 19/11: among a plurality of spatial predictive coding modes - H04N 19/134: characterised by the element, parameter or criterion affecting or controlling the adaptive coding - H04N 19/159: Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction - H04N 19/176: the region being a block, e.g. a macroblock - H04N 19/186: the unit being a colour or a chrominance component - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A block-based decoder is configured to partition a picture of more than one color component and of a color sampling format, according to which each color component is equally sampled, into blocks using a partitioning scheme in which the picture is equally partitioned with respect to each color component. A first color component is decoded in units of the blocks with selecting, for each of intra-predicted first color component blocks, one out of a first set of intra-prediction modes. The first set comprises matrix-based intra prediction modes according to each of which a block inner is predicted by deriving a sample value vector out of neighboring reference samples, computing a matrix-vector product between the sample value vector and a prediction matrix associated with the respective matrix-based intra prediction mode to obtain a prediction vector, and predicting samples in the block inner on the basis of the prediction vector. ## Claims ### Claim 1 A method of reconstructing a picture of more than one color component, the method comprising: partitioning a first color component of the picture into coding tree blocks; decoding a first block of the first color component using matrix-based intra prediction by selecting one out of a first set of intra-prediction modes, the first set comprising matrix-based intra prediction modes according to each of which an inner block is predicted by deriving a sample value vector out of references samples neighboring the inner block, computing a matrix-vector product between the sample value vector and a prediction matrix associated with the respective matrix-based intra prediction mode so as to obtain a prediction vector, and predicting samples in the inner block based on the prediction vector; and decoding a second block of a second color component of the picture, which is co-located with the first block of the first color component using a direct mode, meaning that an intra prediction mode of the second block is derived based on the intra prediction mode selected for the first block of the first color component, the method further comprising determining whether or not a 4:4:4 color sampling format is used in which color components are equally sampled and also determining whether or not a single tree is used for partitioning the first and second color components into coding tree blocks, and if it is determined that both 4:4:4 color sampling and single tree are used, decoding the second block using the matrix-based intra prediction mode selected for the first block of the first color component and if it is determined that either or both of 4:4:4 color sampling and single tree are not used, then decoding the second block using a planar intra prediction mode. ### Claim 2 (depends on 1) The method of claim 1, wherein the matrix-based intra prediction modes comprised by the first set of intra-prediction modes are selected, depending on block dimensions of the first block, so as to be one subset of matrix-based intra prediction modes out of a collection of mutually disjoint subsets of matrix-based intra prediction modes. ### Claim 3 (depends on 1) The method of claim 1, wherein prediction matrices associated with matrix-based intra prediction modes comprised by the first set of intra-prediction modes are of mutually equal size. ### Claim 4 (depends on 1) The method of claim 1, wherein intra prediction modes comprised by the first set of intra-prediction modes, other than the matrix-based intra prediction modes comprised by the first set of intra-prediction modes, comprise a DC mode, a planar mode, and directional modes. ### Claim 5 A video decoder comprising at least one processor and memory, the memory comprising instructions which, when executed by the at least one processor, cause the video decoder to: partition a first color component of a picture of more than one color component into coding tree blocks; decode a first block of the first color component using matrix-based intra prediction by selecting one out of a first set of intra-prediction modes, the first set comprising matrix-based intra prediction modes according to each of which an inner block is predicted by deriving a sample value vector out of references samples neighboring the inner block, computing a matrix-vector product between the sample value vector and a prediction matrix associated with the respective matrix-based intra prediction mode so as to obtain a prediction vector, and predicting samples in the inner block based on the prediction vector; and decode a second block of a second color component of the picture, which is co-located with the first block of the first color component using a direct mode, meaning that an intra prediction mode of the second block is derived based on the intra prediction mode of the first block, the memory further comprising instructions which, when executed by the at least one processor determine whether or not a 4:4:4 color sampling format is used in which color components are equally sampled and also determine whether or not a single tree is used for partitioning the first and second color components into coding tree blocks, and if it is determined that both 4:4:4 color sampling and single tree are used, decode the second block using the matrix-based intra prediction mode selected for the first block of the first color component and if it is determined that either or both of 4:4:4 color sampling and single tree are not used, then decode the second block using a planar intra prediction mode. ### Claim 6 (depends on 5) The video decoder of claim 5, wherein the matrix-based intra prediction modes comprised by the first set of intra-prediction modes are selected, depending on block dimensions of the first block, so as to be one subset of matrix-based intra prediction modes out of a collection of mutually disjoint subsets of matrix-based intra prediction modes. ### Claim 7 (depends on 5) The video decoder of claim 5, wherein prediction matrices associated with matrix-based intra prediction modes comprised by the first set of intra-prediction modes are of mutually equal size and machine-learned. ### Claim 8 (depends on 5) The video decoder of claim 5, wherein intra prediction modes comprised by the first set of intra-prediction modes, other than the matrix-based intra prediction modes comprised by the first set of intra-prediction modes, comprise a DC mode, a planar mode, and directional modes. ### Claim 9 A method of encoding a picture of more than one color component, the method comprising: partitioning a first color component of the picture into coding tree blocks; encoding a first block of the first color component using matrix-based intra prediction by selecting one out of a first set of intra-prediction modes, the first set comprising matrix-based intra prediction modes according to each of which an inner block is predicted by deriving a sample value vector out of references samples neighboring the inner block, computing a matrix-vector product between the sample value vector and a prediction matrix associated with the respective matrix-based intra prediction mode so as to obtain a prediction vector, and predicting samples in the inner block based on the prediction vector; and encoding a second block of a second color component of the picture, which is co-located with the first block of the first color component using a direct mode, meaning that an intra prediction mode of the second block is derived based on the intra prediction mode selected for the first block of the first color component, the method further comprising determining whether or not a 4:4:4 color sampling format is used in which color components are equally sampled and also determining whether or not a single tree is used for partitioning the first and second color components into coding tree blocks, and if it is determined that both 4:4:4 color sampling and single tree are used, encoding the second block using the matrix-based intra prediction mode selected for the first block of the first color component and if it is determined that either or both of 4:4:4 color sampling and single tree are not used, then encoding the second block using a planar intra prediction mode. ### Claim 10 (depends on 9) The method of claim 9, wherein the matrix-based intra prediction modes comprised by the first set of intra-prediction modes are selected, depending on block dimensions of the first block, so as to be one subset of matrix-based intra prediction modes out of a collection of mutually disjoint subsets of matrix-based intra prediction modes. ### Claim 11 (depends on 9) The method of claim 9, wherein prediction matrices associated with matrix-based intra prediction modes comprised by the first set of intra-prediction modes are of mutually equal size. ### Claim 12 (depends on 9) The method of claim 9, wherein intra prediction modes comprised by the first set of intra-prediction modes, other than the matrix-based intra prediction modes comprised by the first set of intra-prediction modes, comprise a DC mode, a planar mode, and directional modes. ### Claim 13 A video encoder comprising at least one processor and memory, the memory comprising instructions which, when executed by the at least one processor, cause the video encoder to: partition a first color component of a picture into coding tree blocks; encode a first block of the first color component using matrix-based intra prediction by selecting one out of a first set of intra-prediction modes, the first set comprising matrix-based intra prediction modes according to each of which an inner block is predicted by deriving a sample value vector out of references samples neighboring the inner block, computing a matrix-vector product between the sample value vector and a prediction matrix associated with the respective matrix-based intra prediction mode so as to obtain a prediction vector, and predicting samples in the inner block based on the prediction vector; and encode a second block of a second color component of the picture, which is co-located with the first block of the first color component using a direct mode, meaning that an intra prediction mode of the second block is derived based on the intra prediction mode selected for the first block of the first color component, the memory further comprising instructions which, when executed by the at least one processor determine whether or not a 4:4:4 color sampling format is used in which color components are equally sampled and also determine whether or not a single tree is used for partitioning the first and second color components into coding tree blocks, and if it is determined that both 4:4:4 color sampling and single tree are used, encode the second block using the matrix-based intra prediction mode selected for the first block of the first color component and if it is determined that either or both of 4:4:4 color sampling and single tree are not used, then encode the second block using a planar intra prediction mode. ### Claim 14 (depends on 13) The video encoder of claim 13, wherein the matrix-based intra prediction modes comprised by the first set of intra-prediction modes are selected, depending on block dimensions of the first block, so as to be one subset of matrix-based intra prediction modes out of a collection of mutually disjoint subsets of matrix-based intra prediction modes. ### Claim 15 (depends on 13) The video encoder of claim 13, wherein prediction matrices associated with matrix-based intra prediction modes comprised by the first set of intra-prediction modes are of mutually equal size and machine-learned. ### Claim 16 (depends on 13) The video encoder of claim 13, wherein intra prediction modes comprised by the first set of intra-prediction modes, other than the matrix-based intra prediction modes comprised by the first set of intra-prediction modes, comprise a DC mode, a planar mode, and directional modes. ### Claim 17 A non-transitory digital storage medium having a computer program stored thereon to perform, when the computer program is run by a computer, a method of video decoding, the method comprising: partitioning a first color component of a picture into coding tree blocks; decoding a first block of the first color component using matrix-based intra prediction by selecting one out of a first set of intra-prediction modes, the first set comprising matrix-based intra prediction modes according to each of which an inner block is predicted by deriving a sample value vector out of references samples neighboring the inner block, computing a matrix-vector product between the sample value vector and a prediction matrix associated with the respective matrix-based intra prediction mode so as to obtain a prediction vector, and predicting samples in the inner block based on the prediction vector; and decoding a second block of a second color component of the picture, which is co-located with the first block of the first color component using a direct mode, meaning that an intra prediction mode of the second block is derived based on the intra prediction mode selected for the first block of the first color component, the method further comprising determining whether or not a 4:4:4 color sampling format is used in which color components are equally sampled and also determining whether or not a single tree is used for partitioning the first and second color components into coding tree blocks, and if it is determined that both 4:4:4 color sampling and single tree are used, decoding the second block using the matrix-based intra prediction mode selected for the first block of the first color component and if it is determined that either or both of 4:4:4 color sampling and single tree are not used, then decoding the second block using a planar intra prediction mode. ### Claim 18 (depends on 17) The non-transitory digital storage medium of claim 17, wherein the matrix-based intra prediction modes comprised by the first set of intra-prediction modes are selected, depending on block dimensions of the first block, so as to be one subset of matrix-based intra prediction modes out of a collection of mutually disjoint subsets of matrix-based intra prediction modes. ### Claim 19 A non-transitory digital storage medium having a computer program stored thereon to perform, when the computer program is run by a computer, a method of video encoding, the method comprising: partitioning a first color component of a picture into coding tree blocks; encoding a first block of the first color component using matrix-based intra prediction by selecting one out of a first set of intra-prediction modes, the first set comprising matrix-based intra prediction modes according to each of which an inner block is predicted by deriving a sample value vector out of references samples neighboring the inner block, computing a matrix-vector product between the sample value vector and a prediction matrix associated with the respective matrix-based intra prediction mode so as to obtain a prediction vector, and predicting samples in the inner block based on the prediction vector; and encoding a second block of a second color component of the picture, which is co-located with the first block of the first color component using a direct mode, meaning that an intra prediction mode of the second block is derived based on the intra prediction mode selected for the first block of the first color component, the method further comprising determining whether or not a 4:4:4 color sampling format is used in which color components are equally sampled and also determining whether or not a single tree is used for partitioning the first and second color components into coding tree blocks, and if it is determined that both 4:4:4 color sampling and single tree are used, encoding the second block using the matrix-based intra prediction mode selected for the first block of the first color component and if it is determined that either or both of 4:4:4 color sampling and single tree are not used, then encoding the second block using a planar intra prediction mode. ### Claim 20 (depends on 19) The non-transitory digital storage medium of claim 19, wherein the matrix-based intra prediction modes comprised by the first set of intra-prediction modes are selected, depending on block dimensions of the first block, so as to be one subset of matrix-based intra prediction modes out of a collection of mutually disjoint subsets of matrix-based intra prediction modes.