A point cloud decoding device according to the present invention including: a geometry information decoding unit that decodes syntax used to indicate the number of layers of a tree in decoding an Octree, wherein the syntax decoded by the geometry information decoding unit has a value equal to or less than a value obtained by adding a predetermined natural number to a maximum node size per slice or per data unit.

A point cloud decoding device according to the present invention including: a geometry information decoding unit that decodes syntax used to indicate the number of layers of a tree in decoding an Octree, wherein the syntax decoded by the geometry information decoding unit has a value equal to or less than a value obtained by adding a predetermined natural number to a maximum node size per slice or per data unit.

A point cloud decoding device 200 according to the present invention includes: a geometry information decoding unit 2010 that decodes a first flag used to indicate whether or not to use a planar mode from a bit stream and recognizes the first flag as having an identical value to a case of failing to use the planar mode in a case of non-inclusion of the first flag in the bit stream.

A point cloud decoding device 200 according to the present invention includes: a geometry information decoding unit 2010 that decodes a first flag used to indicate whether or not to use a planar mode from a bit stream and recognizes the first flag as having an identical value to a case of failing to use the planar mode in a case of non-inclusion of the first flag in the bit stream.

A point cloud decoding device according to the present invention 200 includes: a geometry information decoding unit 2010 that decodes a first flag used to indicate whether or not to carry out inverse quantization of position information from a bit stream; and a tree synthesizing unit 2020 that sets a value of NodeQp to zero in a case where the first flag indicates that the inverse quantization of the position information is not carried out, the NodeQp being used as a quantization parameter for each node.

A point cloud decoding device according to the present invention 200 includes: a geometry information decoding unit 2010 that decodes a first flag used to indicate whether or not to carry out inverse quantization of position information from a bit stream; and a tree synthesizing unit 2020 that sets a value of NodeQp to zero in a case where the first flag indicates that the inverse quantization of the position information is not carried out, the NodeQp being used as a quantization parameter for each node.

A three-dimensional data encoding method includes: calculating a prediction residual that is a difference between geometry information of a three-dimensional point included in point cloud data and a predicted value; quantizing the prediction residual using a first quantization parameter; and generating a bitstream including (i) first information indicating the prediction residual quantized, (ii) second information indicating a second quantization parameter, (iii) third information indicating a first difference between the second quantization parameter and the first quantization parameter, (iv) fourth information indicating a first bit count of the first information, and (v) fifth information indicating a second bit count of the fourth information. When the first difference is not zero, the bitstream further includes sixth information indicating an amount of change in the fifth information that is in accordance with the first difference, and when the first difference is zero, the bitstream does not include the sixth information.

A three-dimensional data encoding method includes: calculating a prediction residual that is a difference between geometry information of a three-dimensional point included in point cloud data and a predicted value; quantizing the prediction residual using a first quantization parameter; and generating a bitstream including (i) first information indicating the prediction residual quantized, (ii) second information indicating a second quantization parameter, (iii) third information indicating a first difference between the second quantization parameter and the first quantization parameter, (iv) fourth information indicating a first bit count of the first information, and (v) fifth information indicating a second bit count of the fourth information. When the first difference is not zero, the bitstream further includes sixth information indicating an amount of change in the fifth information that is in accordance with the first difference, and when the first difference is zero, the bitstream does not include the sixth information.

In a 3D point cloud compression system based on multi-scale structured dictionary learning, a point cloud data partition module outputs a voxel set and a set of blocks of voxels of different scales. A geometric information encoding module outputs an encoded geometric information bit stream. A geometric information decoding module outputs decoded geometric information. An attribute signal encoding module outputs a sparse coding coefficient matrix and a learned multi-scale structured dictionary. An attribute signal compression module outputs a compressed attribute signal bit stream. An attribute signal decoding module outputs decoded attribute signals. A 3D point cloud reconstruction module completes reconstruction. The system is applicable to lossless geometric and lossy attribute compression of point cloud signals. Based on the natural hierarchical partitioning structure of point cloud signals, the system gradually improves the reconstruction quality of high-frequency details in the signals from coarse scale to fine scale, and achieves significant gains.

In a 3D point cloud compression system based on multi-scale structured dictionary learning, a point cloud data partition module outputs a voxel set and a set of blocks of voxels of different scales. A geometric information encoding module outputs an encoded geometric information bit stream. A geometric information decoding module outputs decoded geometric information. An attribute signal encoding module outputs a sparse coding coefficient matrix and a learned multi-scale structured dictionary. An attribute signal compression module outputs a compressed attribute signal bit stream. An attribute signal decoding module outputs decoded attribute signals. A 3D point cloud reconstruction module completes reconstruction. The system is applicable to lossless geometric and lossy attribute compression of point cloud signals. Based on the natural hierarchical partitioning structure of point cloud signals, the system gradually improves the reconstruction quality of high-frequency details in the signals from coarse scale to fine scale, and achieves significant gains.