# US 12,003,767 B2 · Three-dimensional data encoding method, three-dimensional data decoding method, three-dimensional data encoding device, and three-dimensional data decoding device > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,003,767 B2 - Document ID: US12003767B2 - Status: Granted - Kind: B2 - Application number: 17/475,673 - Filed: 2021-09-15 - Published: 2022-01-06 - Granted: 2024-06-04 - Earliest claimed: 2019-03-26 - Assignees: Panasonic Intellectual Property Corporation of America - Applicants: Panasonic Intellectual Property Corporation of America - Inventors: Toshiyasu Sugio; Noritaka Iguchi - Web page: https://patentgrep.com/patent/US12003767B2 ## CPC - G06T 9/004: {Predictors, e.g. intraframe, interframe coding} - H04N 19/597: specially adapted for multi-view video sequence encoding - G06T 9/40: Tree coding, e.g. quadtree, octree - 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 19/13: Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC] - H04N 19/184: the unit being bits, e.g. of the compressed video stream - H04N 19/187: the unit being a scalable video layer - H04N 19/30: using hierarchical techniques, e.g. scalability (H04N19/63 takes precedence) - 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 - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A three-dimensional data encoding method includes: assigning three-dimensional points to M layers, where M is an integer greater than 1; generating respective predicted values of attribute information for the three-dimensional points; encoding the attribute information based on the respective predicted values; and generating a bitstream including layer-number information indicating a number N of lower layers including a bottom-most layer among the M layers, where N is an integer between zero and M, inclusive. In the generating of the predicted values: (i) a predicted value for an upper-layer three-dimensional point in upper layers other than the lower layers among the M layers is generated by a same-layer reference in which another three-dimensional point belonging to a same layer as the upper-layer three-dimensional point is referenced; and (ii) a predicted value for a lower-layer three-dimensional point in lower layers is generated in a condition where the same-layer reference is disabled. ## Claims ### Claim 1 A three-dimensional data encoding method, comprising: assigning three-dimensional points to M Levels of Detail (LoDs), M being an integer greater than 1; generating respective predicted values of attribute information for the three-dimensional points; encoding the attribute information for the three-dimensional points based on the respective predicted values; and generating a bitstream including detail level number information indicating a number N specifying a finest detail level among a first LoD group included in the M LoDs, N being an integer greater than or equal to zero and less than or equal to M, wherein in the generating of the predicted values: (i) a predicted value for a first three-dimensional point belonging to a first LoD included in the first LoD group is generated in a condition where a same detail level reference is enabled to the first three-dimensional point belonging to the first LoD, the first LoD being specified based on the detail level number information, the same detail level reference referencing attribute information of other three-dimensional points included in a same LoD to which the three-dimensional point to be encoded belongs, the first LoD group including one or more LoDs from a coarsest LoD among the M LoDs to the finest detail level among the first LoD group specified by N; and (ii) a predicted value for a second three-dimensional point belonging to a second LoD included in a second LoD group is generated in a condition where the same detail level reference is disabled to the second three-dimensional point belonging to the second LoD included in the second LoD group, the second LoD group including one or more LoDs other than the first LoDs included in the first LoD group among the M LoDs. ### Claim 2 (depends on 1) The three-dimensional data encoding method according to claim 1, wherein the bitstream further includes information indicating a maximum permitted number of LoDs that can be set for the M LoDs. ### Claim 3 (depends on 2) The three-dimensional data encoding method according to claim 2, wherein in the generating of the predicted values, when M is less than the maximum permitted number of LoDs and N is greater than or equal to M, the same detail level reference is permitted for a three-dimensional point belonging to the coarsest LoD among the M LoDs. ### Claim 4 (depends on 1) The three-dimensional data encoding method according to claim 1, wherein in the generating of the predicted values, when N is equal to 0, the same detail level reference is permitted for all of the three-dimensional points. ### Claim 5 (depends on 1) The three-dimensional data encoding method according to claim 1, wherein in the generating of the predicted values, when N is equal to M, the same detail level reference is disabled for all of the three-dimensional points. ### Claim 6 (depends on 1) The three-dimensional data encoding method according to claim 1, wherein three-dimensional points belonging to the first LoD included in the M LoDs are sparser than three-dimensional points belonging to a LoD that is finer than the first LoD. ### Claim 7 (depends on 1) The three-dimensional data encoding method according to claim 1, wherein a distance between three-dimensional points belonging to a first LoD included in the M LoDs is greater than a distance between three-dimensional points belonging to an LoD lower than the first LoD. ### Claim 8 A three-dimensional data decoding method, comprising: obtaining, from a bitstream, (i) encoded attribute information generated by encoding attribute information of three-dimensional points assigned to M Levels of Detail (LoDs), M being an integer greater than 1, and (ii) detail level number information indicating a number N specifying a finest detail level among a first LoD group included in the M LoDs, N being an integer greater than or equal to zero and less than or equal to M; generating respective predicted values of the attribute information for the three-dimensional points; and decoding the encoded attribute information for the three-dimensional points based on the respective predicted values, wherein in the generating of the predicted values: (i) a predicted value for a first three-dimensional point belonging to a first LoD included in the first LoD group is generated in a condition where a same detail level reference is enabled to the first three-dimensional point belonging to the first LoD, the first LoD being specified based on the detail level number information, the same detail level reference referencing attribute information of other three-dimensional points included in a same LoD to which the three-dimensional point to be encoded belongs, the first LoD group including one or more LoDs from a coarsest LoD among the M LoDs to the finest detail level among the first LoD group specified by N; and (ii) a predicted value for a second three-dimensional point belonging to a second LoD included in a second LoD group is generated in a condition where the same detail level reference is disabled to the second three-dimensional point belonging to the second LoD included in the second LoD group, the second LoD group including one or more LoDs other than the first LoDs included in the first LoD group among the M LoDs. ### Claim 9 A three-dimensional data encoding device, comprising: a processor; and memory, wherein using the memory, the processor: assigns three-dimensional points to M Levels of Detail (LoDs), M being an integer greater than 1; generates respective predicted values of attribute information for the three-dimensional points; encodes the attribute information for the three-dimensional points based on the respective predicted values; and generates a bitstream including detail level number information indicating a number N specifying a finest detail level among a first LoD group included in the M LoDs, N being an integer greater than or equal to zero and less than or equal to M, in the generating of the predicted values, the processor: (i) generates a predicted value for a first three-dimensional point belonging to a first LoD included in the first LoD group is generated in a condition where a same detail level reference is enabled to the first three-dimensional point belonging to the first LoD, the first LoD being specified based on the detail level number information, the same detail level reference referencing attribute information of other three-dimensional points included in a same LoD to which the three-dimensional point to be encoded belongs, the first LoD group including one or more LoDs from a coarsest LoD among the M LoDs to the finest detail level among the first LoD group specified by N; and (ii) generates a predicted value for a second three-dimensional point belonging to a second LoD included in a second LoD group is generated in a condition where the same detail level reference is disabled to the second three-dimensional point belonging to the second LoD included in the second LoD group, the second LoD group including one or more LoDs other than the first LoDs included in the first LoD group among the M LoDs. ### Claim 10 A three-dimensional data decoding device, comprising: a processor; and memory, wherein using the memory, the processor: obtains, from a bitstream, (i) encoded attribute information generated by encoding attribute information of three-dimensional points assigned to M Levels of Detail (LoDs), M being an integer greater than 1, and (ii) detail level number information indicating a number N specifying a finest detail level among a first LoD group included in the M LoDs, N being an integer greater than or equal to zero and less than or equal to M; generates respective predicted values of the encoded attribute information for the three-dimensional points; and decodes the encoded attribute information for the three-dimensional points based on the respective predicted values, in the generating of the predicted values, the processor: (i) generates a predicted value for a first three-dimensional point belonging to a first LoD included in the first LoD group is generated in a condition where a same detail level reference is enabled to the first three-dimensional point belonging to the first LoD, the first LoD being specified based on the detail level number information, the same detail level reference referencing attribute information of other three-dimensional points included in a same LoD to which the three-dimensional point to be encoded belongs, the first LoD group including one or more LoDs from a coarsest LoD among the M LoDs to the finest detail level among the first LoD group specified by N; and (ii) generates a predicted value for a second three-dimensional point belonging to a second LoD included in a second LoD group is generated in a condition where the same detail level reference is disabled to the second three-dimensional point belonging to the second LoD included in the second LoD group, the second LoD group including one or more LoDs other than the first LoDs included in the first LoD group among the M LoDs.