In-tree geometry quantization of point clouds

A method comprising:

• receiving, by a computer system, a data structure representing a plurality of points in a point cloud in three-dimensional space;
• decoding, by the computer system, the data structure to determine one or more sequences of bits, wherein each of the sequences of bits represents a position of one of the points in one of the dimensions of the three-dimensional space, and wherein the data structure is decoded according to a tree data structure;
• partitioning, by the computer system, at least one of the sequences of bits into a first portion of bits and a second portion of bits,
• scaling, by the computer system, each of the second portions of bits according to a quantization step size, wherein the quantization step size is determined according to an exponential function having a quantization parameter value as an input and the quantization step size as an output;
• generating, by the computer system, a reconstructed point position of at least one of the points by combining, for each of the at least one of the points, the first portion of bits and the scaled second portion of bits corresponding to that point;
• and
• outputting, by the computer system, the reconstructed point position of the least one of the points.

A device comprising:

• one or more processors;
• and
• memory storing instructions that when executed by the one or more processors, cause the one or more processors to perform operations comprising:
• receiving a data structure representing a plurality of points in a point cloud in three-dimensional space;
• decoding the data structure to determine one or more sequences of bits, wherein each of the sequences of bits represents a position of one of the points in one of the dimensions of the three-dimensional space, and wherein the data structure is decoded according to a tree data structure;
• partitioning at least one of the sequences of bits into a first portion of bits and a second portion of bits;
• scaling each of the second portions of bits according to a quantization step size, wherein the quantization step size is determined according to an exponential function having a quantization parameter value as an input and the quantization step size as an output;
• generating a reconstructed point position of at least one of the points by combining, for each of the at least one of the points, the first portion of bits and the scaled second portion of bits corresponding to that point;
• and
• outputting the reconstructed point position of the least one of the points.

A non-transitory, computer-readable storage medium having instructions stored thereon, that when executed by one or more processors, cause the one or more processors to perform operations comprising:

• receiving a data structure representing a plurality of points in a point cloud in three-dimensional space;
• decoding the data structure to determine one or more sequences of bits, wherein each of the sequences of bits represents a position of one of the points in one of the dimensions of the three-dimensional space, and wherein the data structure is decoded according to a tree data structure;
• partitioning at least one of the sequences of bits into a first portion of bits and a second portion of bits;
• scaling each of the second portions of bits according to a quantization step size, wherein the quantization step size is determined according to an exponential function having a quantization parameter value as an input and the quantization step size as an output;
• generating a reconstructed point position of at least one of the points by combining, for each of the at least one of the points, the first portion of bits and the scaled second portion of bits corresponding to that point;
• and
• outputting the reconstructed point position of the least one of the points.