Guangdong Oppo Mobile Telecommunications Corp., Ltd.

Zexing Sun

Fuzheng Yang

Shuai Wan

Junyan Huo

Yanzhuo Ma

Wei Zhang

Sujun Zhang

Lihui Yang

Na Dai

Point cloud coding methods, an encoder, and a decoder are provided. The method includes the following. Morton codes for a point cloud are obtained according to geometry information of the point cloud after geometry decoding is completed and the geometry information is reconstructed. A value of K is determined according to a distribution of the point cloud, where K is a positive integer, where K is a positive integer. The point cloud is determined as an input point cloud of an iteration. Morton codes for an input point cloud of an i-th iteration are right shifted by K bits according to a number of points in the input point cloud of the i-th iteration. An i-th level of detail (LOD) layer is formed with a point selected after the Morton codes are shifted. Color information is decoded according to the LOD layer.

Point cloud coding methods, encoder, and decoder

The implementation of the present disclosure provides a point cloud partitioning method and device, and a computer-readable storage medium, including: when performing stripe division along the longest side, adjusting an initial partitioning position or determining the stripe division length according to the size of a preset block, to obtain a stripe division position, wherein, the length of the first n−1 stripes along the longest side is an integer multiple of the side length of the preset block, n is the number of divided stripes, and n is a positive integer greater than or equal to 2.

Point cloud segmentation method and device, and computer readable storage medium

Described is a nearest neighbor search method. The method is applicable in a device and comprises: acquiring a Morton code set of point cloud data to be searched; stratifying the point cloud data on the basis of the Morton code set and of a first distance threshold to produce current stratum data; shifting to the right by a first preset digit Morton codes of prediction data other than the stratified data in the point cloud data to produce a corresponding first parent node set; determining, on the basis of Morton codes of the current stratum data, a neighbor area satisfying a criterion in the first parent node set; and determining a nearest neighboring point set of the current stratum data in the neighboring area. Also provided in an exemplary embodiment of the present application are a device and a computer storage medium.

Nearest neighbor search method, apparatus, device, and storage medium

Disclosed are a partitioning method, an encoder, a decoder and a computer storage medium. The method includes: determining location information of a point of a point cloud to be partitioned; when i is less than or equal to M−1, determining right-shift number Nof ith LOD layer in the point cloud, M representing a preset maximum quantity of layers for LOD partitioning; for the ith LOD layer, shifting location information of the point rightwards by N-digit, performing storing in a preset storage area based on right-shifted location information; determining location information of a parent point corresponding to a current point in the ith LOD layer; according to determined location information of the parent point, searching the preset storage area for a neighbor point of the parent point; partitioning the current point into an (i+1)th LOD layer, or the neighbor point into the ith LOD layer.

Partitioning method, encoder, decoder and computer storage medium

Provided in the embodiments of the present application are a transform method, inverse transform method, encoder, decoder and storage medium. The method comprises: determining a two-dimensional Morton code of an encoding point in a preset space of an encoding point cloud, wherein the two-dimensional Morton code is a Morton code corresponding to two-dimensional coordinates of the encoding point obtained by removing one of three-dimensional coordinates of the encoding point; performing right shift of the two-dimensional Morton code by K binary digits to determine point sets respectively corresponding to three-dimensional coordinate components; determining, from the point sets, point quantities of maximum included encoding points respectively corresponding to the three-dimensional coordinate components; determining, based on the point quantities, a transform order corresponding to the preset space; performing RAHT based on the transform order; and signaling the transform order and an attribute encoding result obtained after RAHT into a bitstream.

Transform method, inverse transform method, encoder, decoder and storage medium

A point cloud model reconstruction method, an encoder, and a decoder are provided. The method includes the following. An i-th segmentation position of an i-th three-dimensional slice on a longest edge in a point cloud space is obtained, where 1&#x2264;i&#x2264;n&#x2212;1 and n is a total number of three-dimensional slices to-be-processed. An i-th cube block set nearest to the i-th segmentation position in the i-th three-dimensional slice is determined based on the i-th segmentation position and a preset cube block length. An i-th vertex position with a cube block in the i-th cube block set is determined along the longest edge, where the i-th vertex position represents a position of a vertex on an edge which is away from a start position of the cube block by the preset cube block length. A point cloud model is reconstructed based on the i-th vertex position.

Point cloud model reconstruction method, encoder, and decoder

Point cloud coding methods, an encoder, and a decoder are provided. The method includes the following. Morton codes for a point cloud are obtained according to geometry information of the point cloud after geometry decoding is completed and the geometry information is reconstructed. The point cloud is determined as an input point cloud of an iteration. Morton codes for an input point cloud of an i-th iteration are right shifted by K bits according to a number of points in the input point cloud of the i-th iteration, where K is a positive integer. A point is selected from points with a same value of Morton codes after right shifted by K bits. An i-th level of detail (LOD) layer is formed with the point selected. Color information is decoded according to the LOD layer.

Implementations of this application provide point cloud partition methods, an encoder, and a decoder. The method includes the following. Morton codes for a point cloud are obtained according to geometry information of the point cloud. The point cloud is determined as an input point cloud of an iteration. Morton codes for an input point cloud of an i-th iteration are right shifted by K bits according to a number of points in the input point cloud of the i-th iteration, where K is a positive integer. A point is selected from points with a same value of Morton codes after right shifted by K bits. An i-th level of detail (LOD) layer is formed with the point selected.

Growing community of inventors

Guangdong, China

Zexing Sun