WO2022116117A1

WO2022116117A1 - Prediction method, encoder, decoder and computer storage medium

Info

Publication number: WO2022116117A1
Application number: PCT/CN2020/133704
Authority: WO
Inventors: 杨付正; 孙泽星; 代娜; 万帅; 霍俊彦; 马彦卓
Original assignee: Oppo广东移动通信有限公司
Priority date: 2020-12-03
Filing date: 2020-12-03
Publication date: 2022-06-09
Also published as: CN116547968A; CN117793372A

Abstract

Disclosed are a prediction method, an encoder, a decoder and a computer storage medium. The method is applied to an encoder. The method comprises: determining N neighbor nodes to be searched for that are of the current node; if there is a coplanar neighbor node that is coplanar with the current node and has been encoded among said N neighbor nodes, acquiring a first attribute reconstruction value of the coplanar neighbor node; if there is a collinear neighbor node that is collinear with the current node and has been encoded among said N neighbor nodes, acquiring a second attribute reconstruction value of the collinear neighbor node; and on the basis of the first attribute reconstruction value and/or the second attribute reconstruction value, determining an attribute prediction value of the current node.

Description

Prediction method, encoder, decoder, and computer storage medium

technical field

The embodiments of the present application relate to the technical field of encoding and decoding, and in particular, to a prediction method, an encoder, a decoder, and a computer storage medium.

Background technique

In the Point Cloud Reference Model (PCRM) coding framework based on the Audio Video Coding Standard (AVS), the geometric information of the point cloud and the attribute information corresponding to each point cloud are encoded separately. After the geometric encoding is completed, the geometric information will be reconstructed, and the encoding of the attribute information will depend on the reconstructed geometric information. Among them, the attribute information coding is mainly for the coding of color information, so as to transform the color information from the spatial domain to the frequency domain to obtain high-frequency coefficients and low-frequency coefficients, and finally perform quantization and entropy coding on the coefficients to generate a code stream.

However, the current attribute prediction does not fully utilize the spatial correlation of point clouds, which reduces the encoding and decoding efficiency of point clouds.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a prediction method, an encoder, a decoder, and a computer storage medium, which can make full use of the spatial correlation of point clouds, thereby reducing the code rate and improving encoding and decoding efficiency.

The technical solutions of the embodiments of the present application can be implemented as follows:

In a first aspect, an embodiment of the present application provides a prediction method, which is applied to an encoder, and the method includes:

Determine the N neighbor nodes to be searched for the current node;

If there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched, acquiring the first attribute reconstruction value of the coplanar neighbor node;

If there is a collinear neighbor node that is collinear with the current node and has been encoded in the N neighbor nodes to be searched, acquiring the second attribute reconstruction value of the collinear neighbor node;

An attribute prediction value of the current node is determined based on the first attribute reconstruction value and/or the second attribute reconstruction value.

In a second aspect, an embodiment of the present application provides a prediction method, which is applied to a decoder, and the method includes:

Determine the N neighbor nodes to be searched for the current node;

If there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched, acquiring the reconstruction value of the first attribute of the coplanar neighbor node;

If there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched, acquiring the second attribute reconstruction value of the collinear neighbor node;

In a third aspect, an embodiment of the present application provides an encoder, the encoder includes a first determination unit, a first judgment unit, and a first prediction unit; wherein,

The first determining unit is configured to determine the N neighbor nodes to be searched for the current node;

The first judgment unit is configured to obtain a first attribute reconstruction value of the coplanar neighbor node if there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched ;

The first judging unit is further configured to obtain the second attribute reconstruction of the collinear neighbor node if there is a collinear neighbor node that is collinear with the current node and has been encoded in the N neighbor nodes to be searched value;

The first prediction unit is configured to determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.

In a fourth aspect, an embodiment of the present application provides an encoder, where the encoder includes a first memory and a first processor; wherein,

a first memory for storing a computer program executable on the first processor;

The first processor is configured to execute the method according to the first aspect when running the computer program.

In a fifth aspect, an embodiment of the present application provides a decoder, the decoder includes a second determination unit, a second judgment unit, and a second prediction unit; wherein,

The second determining unit is configured to determine the N neighbor nodes to be searched for the current node;

The second judgment unit is configured to obtain a first attribute reconstruction value of the coplanar neighbor node if there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched ;

The second judgment unit is further configured to obtain the second attribute reconstruction of the collinear neighbor node if there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched value;

The second prediction unit is configured to determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.

In a sixth aspect, an embodiment of the present application provides a decoder, the decoder includes a second memory and a second processor; wherein,

a second memory for storing a computer program executable on the second processor;

The second processor is configured to execute the method according to the second aspect when running the computer program.

In a seventh aspect, an embodiment of the present application provides a computer storage medium, where the computer storage medium stores a computer program, and when the computer program is executed by the first processor, the method described in the first aspect is implemented, or the computer program is executed by the second processor. The processor implements the method as described in the second aspect when executed.

The embodiments of the present application provide a prediction method, an encoder, a decoder, and a computer storage medium. On the encoder side, N neighbor nodes to be searched for the current node are determined; If the current node is coplanar and has been coded coplanar neighbor nodes, the first attribute reconstruction value of the coplanar neighbor node is obtained; if the N neighbor nodes to be searched are collinear with the current node and already the encoded collinear neighbor node, obtain the second attribute reconstruction value of the collinear neighbor node; based on the first attribute reconstruction value and/or the second attribute reconstruction value, determine the attribute prediction value of the current node . On the decoder side, by determining the N neighbor nodes to be searched of the current node; if there is a coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, obtain the coplanar neighbor node. The first attribute reconstruction value of the neighbor node; if there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the collinear neighbor node ; determining an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value. In this way, when the attribute prediction is performed according to the geometric spatial relationship between the current node and the neighbor nodes, due to the expansion of the neighbor search range, the spatial correlation of the point cloud can be fully utilized, so that the prediction residual is smaller, and the code rate can be reduced. Thus, the encoding and decoding efficiency is improved.

Description of drawings

1A is a schematic diagram of the frame composition of a point cloud encoder provided by the related art;

1B is a schematic diagram of the frame composition of a point cloud decoder provided by the related art;

2 is a schematic diagram of the spatial relationship between a current node and a neighbor node provided by the related art;

3A is a schematic diagram of a Morton code relationship between neighbor nodes that are coplanar with the current node and have been encoded and decoded within a preset neighbor range provided by the related art;

3B is a schematic diagram of the Morton code relationship between neighbor nodes that are collinear with the current node and have been encoded and decoded within a preset neighbor range provided by the related art;

4 is a schematic flowchart of a prediction method provided by an embodiment of the present application;

5 is a schematic flowchart of another prediction method provided by an embodiment of the present application;

6A is a schematic diagram of a neighbor search range provided by the related art;

6B is a schematic diagram of a neighbor search range provided by an embodiment of the present application;

7 is a schematic flowchart of another prediction method provided by an embodiment of the present application;

FIG. 8 is a schematic diagram of the composition and structure of an encoder according to an embodiment of the present application;

9 is a schematic diagram of a specific hardware structure of an encoder provided by an embodiment of the present application;

FIG. 10 is a schematic diagram of the composition and structure of a decoder provided by an embodiment of the application;

FIG. 11 is a schematic diagram of a specific hardware structure of a decoder provided by an embodiment of the present application.

Detailed ways

In order to have a more detailed understanding of the features and technical contents of the embodiments of the present application, the implementation of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is understood that "some embodiments" can be the same or a different subset of all possible embodiments, and Can be combined with each other without conflict.

It should be pointed out that the term "first\second\third" involved in the embodiments of the present application is only to distinguish similar objects, and does not represent a specific ordering of objects. It is understandable that "first\second\third" "Where permitted, the specific order or sequence may be interchanged to enable the embodiments of the application described herein to be practiced in sequences other than those illustrated or described herein.

Before the embodiments of the present application are described in further detail, the nouns and terms involved in the embodiments of the present application are described. The nouns and terms involved in the embodiments of the present application are applicable to the following explanations: point cloud compression (Point Cloud Compression, PCC), Point Cloud Reference Model (PCRM), Morton code (Morton), bounding box, octree, Red-Green-Blue (RGB), Luminance-Chrominance (YUV), Look Up Table (LUT), Peak Signal to Noise Ratio (PSNR), Audio Video coding Standard (AVS).

Point cloud is a three-dimensional representation of the surface of an object. Through photoelectric radar, lidar, laser scanner, multi-view camera and other acquisition equipment, the point cloud (data) on the surface of the object can be collected.

Specifically, a point cloud refers to a collection of massive three-dimensional points, and the points in the point cloud may include point location information and point attribute information. For example, the position information of the point may be three-dimensional coordinate information of the point. The position information of the point may also be referred to as the geometric information of the point. For example, the attribute information of the points may include color information and/or reflectivity, among others. For example, the color information may be information in any color space. For example, the color information may be RGB information. Among them, R represents red (Red, R), G represents green (Green, G), and B represents blue (Blue, B). For another example, the color information may be luminance chrominance (YCbCr, YUV) information. Among them, Y represents brightness (Luma), Cb (U) represents blue color difference, and Cr (V) represents red color difference.

According to the point cloud obtained by the principle of laser measurement, the points in the point cloud may include the three-dimensional coordinate information of the point and the laser reflection intensity (reflectance) of the point. For another example, according to the point cloud obtained by the principle of photogrammetry, the points in the point cloud may include three-dimensional coordinate information of the point and color information of the point. For another example, a point cloud is obtained by combining the principles of laser measurement and photogrammetry, and the points in the point cloud may include three-dimensional coordinate information of the point, laser reflection intensity (reflectance) of the point, and color information of the point.

Point clouds can be divided into:

The first type of static point cloud: that is, the object is static, and the device that obtains the point cloud is also static;

The second type of dynamic point cloud: the object is moving, but the device that obtains the point cloud is stationary;

The third type of dynamic point cloud acquisition: the device that acquires the point cloud is moving.

For example, point clouds are divided into two categories according to their use:

Category 1: Machine perception point cloud, which can be used in scenarios such as autonomous navigation systems, real-time inspection systems, geographic information systems, visual sorting robots, and rescue and relief robots;

Category 2: Human eye perception point cloud, which can be used in point cloud application scenarios such as digital cultural heritage, free viewpoint broadcasting, 3D immersive communication, and 3D immersive interaction.

Since the point cloud is a collection of massive points, storing the point cloud will not only consume a lot of memory, but also is not conducive to transmission, and there is no such a large bandwidth to support the direct transmission of the point cloud at the network layer without compression. cloud for compression.

It should be understood that the point cloud compression can generally adopt the way of compressing geometric information and attribute information separately.

In the framework of the PCRM encoder of AVS as shown in Figure 1A, the input point cloud can be divided into geometric information and attribute information corresponding to each point, and the two are encoded separately. First, coordinate transformation (ie, coordinate translation) is performed on the geometric information, so that all point clouds are located in a bounding box, and then coordinate quantization is performed. This step of quantization mainly plays the role of scaling. Then according to the Morton code, the geometric octree is constructed from the root node (level 0) according to breadth-first. Use the octree to divide the bounding box into 8 sub-cubes equally, and record the occupancy information of each sub-cube (wherein, if the occupancy information is 1, then the sub-cube is non-empty; if the occupancy information is 0, then The subcube is empty). Then, continue to divide the non-empty sub-cubes into eight equal parts, until the leaf nodes obtained by dividing are 1×1×1 unit cubes, and stop dividing. Geometric entropy encoding is performed on the occupancy information to generate a binary geometric code stream.

After the geometry encoding is completed, the geometry information is reconstructed (ie, octree reconstruction) and used as additional information to guide attribute encoding. At present, attribute encoding is mainly carried out for color information. First, the color information is spatially transformed, specifically from the RGB color space to the YUV color space. Then, the point cloud is recolored (ie, attribute interpolation) using the reconstructed geometric information, so that the unencoded attribute information corresponds to the reconstructed geometric information.

Next, attribute prediction is performed. First, the point cloud is reordered based on the Morton code to generate a point cloud sequence that can be used for point cloud attribute prediction. Then, the nearest neighbor node is searched using the geometric space relationship, and the reconstructed attribute value of the found nearest neighbor node is used. Predict the attributes of the current node to be predicted to obtain the predicted attribute value. The attribute prediction residual is obtained by differentiating the real attribute value and the predicted attribute value, the residual quantization of the attribute prediction residual is carried out, and then the binary attribute code stream is obtained by inputting the entropy coding engine.

As shown in FIG. 1B , in the framework of the PCRM decoder of AVS, after obtaining the binary code stream, the geometric code stream and the attribute code stream in the binary code stream are independently decoded respectively. When decoding the geometric code stream, first perform entropy decoding on the geometric code stream to obtain the geometric information of each node; then reconstruct the octree in the same way as the geometric entropy encoding, and then perform Through inverse coordinate quantization and inverse coordinate translation, the decoded geometric information can be obtained, that is, the geometric position of the point cloud can be determined. When decoding the attribute code stream, the decoding geometry information needs to be used as additional information to guide the attribute decoding. First, perform entropy decoding on the attribute code stream to obtain the quantized residual information; then perform inverse quantization to obtain the point cloud residual; in the process of attribute reconstruction, the current node to be predicted can be obtained in the same way as attribute encoding. The attribute prediction value of , and then adding the attribute prediction value and the residual value, the YUV attribute value of the current node to be predicted can be recovered; finally, the attribute information of the point cloud can be obtained through inverse space transformation.

In the current related art, the Morton code can be used to find the spatial neighbor nodes of the current node, and then the attribute prediction of the current node is performed according to the found neighbor nodes. Exemplarily, see FIG. 2 , which shows a schematic diagram of a spatial relationship between a current node and a neighbor node provided by the related art. As shown in Figure 2, the block represented by the bold line is the current node, and its corresponding preset neighborhood range is 3×3×3; where X represents the horizontal coordinate axis, Y represents the vertical coordinate axis, and Z represents z Axis. Here, the total number of neighbor nodes of the current node is 26; among them, the number of neighbor nodes that are coplanar with the current node is 6, the number of neighbor nodes that are collinear with the current node is 12, and the number of neighbor nodes that are co-located with the current node. There are 8. It should be noted that the preset neighborhood range is not limited to 3 × 3 × 3, but can also be 4 × 4 × 4, 5 × 5 × 5, 6 × 6 × 6, or even larger than 6 × The size of 6×6, etc., is not specifically limited in the embodiments of the present application.

With reference to the example of FIG. 2 , it is considered that the search range of the neighbor node is a preset neighborhood range with a size of 3×3×3 of the current node to be encoded. First, the Morton code of the current point node is used to obtain the block with the smallest Morton code value in the neighborhood, and this block is used as the reference block, and then the reference block is used to find the coded code that is coplanar and collinear with the current point node to be coded. Decode neighbor nodes. The Morton code relationship between neighboring nodes that are coplanar with the current node to be encoded and have been encoded/decoded (may be referred to as “encoded and decoded”) within the preset neighborhood range is shown in FIG. 3A . The Morton code relationship between neighbor nodes that are collinear with the current node and have been encoded and decoded within the domain is shown in Figure 3B. Here, in FIG. 3A, the block filled with 7 represents the current node, and the block filled with 3, 5, and 6 is the neighbor node that is coplanar with the current node. In FIG. 3B, the block filled with 7 still represents the current node, and the block filled with 1, 2, and 4 is the neighbor node that is collinear with the current node.

In this way, the reference block is used to search for N neighboring nodes that are coplanar and collinear with the current node and have been encoded and decoded (when the preset neighborhood range is 3×3×3, the value of N is less than or equal to 6 ), and then use these N neighbor nodes to predict the attributes of the current node. If no coplanar or collinear neighbor nodes are found, the current node will be predicted by the reconstructed attribute value of the previous encoded and decoded node whose Morton code is smaller than the current node.

As shown in Figure 3A and Figure 3B, the 6 neighbor nodes filled with 1, 2, 3, 4, 5, and 6 must complete the attribute encoding and decoding before the current node, but in addition to these neighbor nodes, they can also include some Collinear neighbor nodes may also complete attribute encoding and decoding before the current node. However, in the related art, currently only three coplanar and three collinear neighbor nodes as shown in FIG. 3A and FIG. 3B are considered, so some coded and decoded collinear neighbor nodes will be missed. Then, after the six coplanar or collinear neighbor nodes cannot be found, it is spatially inaccurate to directly use the reconstructed attribute value of the previously encoded and decoded node whose Morton code is smaller than the current node as the predicted attribute value. , because Morton codes have periodic transition points, even if Morton codes are adjacent, the spatial position cannot be guaranteed to be adjacent. Therefore, for attribute prediction based on geometric spatial relationship, the spatial correlation of adjacent nodes is not fully utilized, thus reducing the encoding and decoding efficiency of point clouds.

An embodiment of the present application provides a prediction method. The basic idea of the method is: by determining N neighbor nodes to be searched for the current node; Coplanar neighbor node that has been encoded/decoded, obtain the first attribute reconstruction value of the coplanar neighbor node; For a collinear neighbor node, the second attribute reconstruction value of the collinear neighbor node is obtained; based on the first attribute reconstruction value and/or the second attribute reconstruction value, the attribute prediction value of the current node is determined. In this way, both the encoder and the decoder can make full use of the spatial correlation of the point cloud due to the expansion of the neighbor search range when making attribute predictions based on the geometric spatial relationship between the current node and the neighbor nodes, making the prediction residual more efficient. small, so that the code rate can be reduced, and the encoding and decoding efficiency can be improved.

The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

In an embodiment of the present application, the prediction method provided in the embodiment of the present application is applied to a video encoding device, that is, a point cloud encoder (or referred to as a PCRM encoder), and the embodiment of the present application may also be referred to as an encoder for short. The functions implemented by the method can be implemented by the first processor in the encoder calling a computer program. Of course, the computer program can be stored in the first memory. It can be seen that the encoder includes at least a first processor and a first memory.

Referring to FIG. 4 , it shows a schematic flowchart of a prediction method provided by an embodiment of the present application. As shown in Figure 4, the method may include:

S401: Determine N neighbor nodes to be searched for the current node.

It should be noted that, in a point cloud, a point can be all points in the point cloud, or some points in the point cloud, and these points are relatively concentrated in space. Here, the current node may also be referred to as the current point or the current block, and specifically refers to the point or block currently to be encoded in the point cloud.

It should also be noted that, based on the framework of the PCRM encoder shown in FIG. 1A , the method in the embodiment of the present application is mainly applied to the “attribute prediction” part, for the current related art based on the neighbor search range (also referred to as “neighborhood”). The attribute prediction of “domain search range” or “neighbor range to be searched”) is optimized, and the utilization of spatial correlation is improved by expanding the neighbor search range.

It can be understood that the N neighbor nodes to be searched are selected from the neighbor information of the current node. In some embodiments, the determining the N neighbor nodes to be searched for the current node may include:

Perform spatial division on the point cloud to obtain at least one point;

determining the neighbor information of the current node from the at least one point based on the geometric position of the current node;

From the neighbor information of the current node, the N neighbor nodes to be searched are determined.

In this embodiment of the present application, the neighbor information of the current node may include: six neighbor nodes coplanar with the current node, twelve neighbor nodes colinear with the current node, and The eight neighbor nodes of the point.

That is to say, based on the preset neighborhood range of size 3×3×3 shown in FIG. 2 , the number of neighbor nodes of the current node is 26 in total. Among them, the number of neighbor nodes that are coplanar with the current node is 6, the number of neighbor nodes that are collinear with the current node is 12, and the number of neighbor nodes that are co-located with the current node is 8. The N neighbor nodes to be searched in the embodiments of the present application are determined from the 26 neighbor nodes.

Further, the determining the N neighbor nodes to be searched from the neighbor information of the current node may include:

The three neighbor nodes that must have been encoded in the six neighbor nodes that are coplanar with the current node are determined as three neighbor nodes to be searched, and the twelve neighbor nodes that are collinear with the current node must be encoded. Three neighbor nodes are determined as three neighbor nodes to be searched;

Based on the geometric space relationship between the current node and the neighbor nodes, n neighbor nodes that may have been encoded in the remaining nine neighbor nodes collinear with the current node are determined as the n neighbor nodes to be searched, where n is greater than an integer of 0 and less than 9;

The obtained (6+n) neighbor nodes to be searched are determined as the N neighbor nodes to be searched.

It should be noted that, among the 26 neighbor nodes, the current related technology only constitutes a neighbor search range of six neighbor nodes that can be determined to be encoded. Specifically, as shown in FIG. 3A , among the six neighbor nodes coplanar with the current node, there are only three neighbor nodes that are coplanar with the left side, front side and lower side of the current node (respectively with 3, 5, 6 padding) is able to determine that it must have been encoded. As shown in Figure 3B, among the twelve neighbor nodes that are collinear with the current node, there are only three neighbor nodes that are collinear with the left, front, and bottom directions of the current node (filled with 1, 2, and 4, respectively) It is possible to be sure that it must have been encoded.

In the embodiment of the present application, in addition to the fact that the six neighbor nodes that have been encoded can be determined, the embodiment of the present application expands the neighbor search range. That is to say, based on the geometric space relationship between the current node and the neighbor nodes, the n neighbor nodes that may have been encoded in the remaining nine neighbor nodes that are collinear with the current node are determined as the n neighbor nodes to be searched, so that the expansion The latter neighbor search range includes (6+n) neighbor nodes to be searched. Here, n is an integer greater than 0 and less than 9, that is, N is an integer greater than 6.

It should also be noted that in the process of expanding the neighbor search range, since the remaining three neighbor nodes coplanar with the current node can be determined to have not been coded, they are not considered; Neighbor nodes can also determine that they must not have been encoded, so they are not considered; and three of the remaining nine neighbor nodes that are collinear with the current node can determine that they must not have been encoded. Therefore, in this application In a specific example of the embodiment, n is an integer greater than 0 and less than or equal to 6.

In another specific example of the embodiment of the present application, N is an integer greater than 6 and less than or equal to 12. In another specific example of the embodiment of the present application, the value of N may be 12.

Further, in some embodiments, the method may further include: writing the value of N into the code stream.

In this way, since the encoder writes the value of N into the code stream, the subsequent decoder can directly obtain the number of neighbor nodes to be searched (the value of N) by parsing the code stream, so that the neighbor search ranges of the encoder and decoder are consistent .

It should also be understood that after the N neighbor nodes to be searched of the current node are determined, the Morton codes of the N neighbor nodes to be searched need to be further determined. In some embodiments, after determining the N neighbor nodes to be searched for the current node, the method may further include:

Determine the reference point of the current node;

Perform difference calculation on the coordinate information between the reference point and the N neighbor nodes to be searched to obtain N coordinate difference values;

calculating the Morton codes of the N coordinate differences, and storing the Morton codes of the N coordinate differences in a preset look-up table;

Morton codes of the N neighbor nodes to be searched are determined according to the preset lookup table.

It should be noted that the preset look-up table is a Morton code for storing N coordinate difference values. The N coordinate difference values are obtained by calculating the difference value of the coordinate information between the reference point of the current node and the N neighbor nodes to be searched.

For the reference point of the current node, in some embodiments, the determining the reference point of the current node may include:

calculating the Morton code of the current node;

According to the Morton code of the current node, determine the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node;

The neighbor node corresponding to the minimum Morton code is determined as the reference point.

Further, determining the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node according to the Morton code of the current node may include:

Determine the coordinate information of the current node according to the Morton code of the current node;

determining the Morton code of at least one neighbor node in the neighbor information of the current node based on the coordinate information of the current node;

A minimum Morton code is selected from Morton codes of the at least one neighbor node to obtain a neighbor node corresponding to the minimum Morton code.

In the embodiment of the present application, the Morton code corresponding to each point in the point cloud can be obtained by using the coordinate information of the node (or simply referred to as "point"). Here, Morton code can convert multi-dimensional data into one-dimensional data encoding, and Morton code can also be called z-order code, because its encoding order is according to the spatial z order. The specific method for calculating the Morton code is described as follows. For the three-dimensional coordinates (x, y, z) represented by a d-bit binary number for each component, the three components are represented by the following:

Wherein, x _l , y _l , z _l ∈ {0,1} are the binary values corresponding to the highest bit (l=1) to the lowest bit (l=d) of x, y, and z, respectively. The Morton code M starts from the highest bit for x, y, and z, and arranges x _l , y _l , z _l to the lowest bit in turn. The calculation formula of M is as follows:

Among them, m _l′ ∈ {0, 1} are the values from the highest order (l′=1) to the lowest order (l′=3d) of M, respectively. After obtaining the Morton code M of each point in the point cloud, the points in the point cloud are arranged in the order of Morton code from small to large, and the weight w of each point is set to 1. Represented in computer language, similar to the combination of z|(y<<1)|(x<<2).

It should be noted that the Morton code of the current node is first calculated, and then the Morton code of the reference point is located according to the Morton code of the current node. Here, the Morton code of the reference point is the smallest Morton code among the neighbor nodes of the current node. Exemplarily, if the three-dimensional coordinates of the current node are (x, y, z), then the three-dimensional coordinates of the reference point are (x-1, y-1, z-1), and the calculated Morton code is the smallest. of.

After the reference point is obtained, the coordinate difference (offset) between the reference point and the N neighbor nodes to be searched can be calculated, and the N coordinate difference values can be converted into N Morton codes, and then the N Morton codes Stored in a preset lookup table.

Further, in some embodiments, the determining the Morton codes of the N neighbor nodes to be searched according to the preset lookup table may include:

Determine the Morton code of the reference point;

From the preset lookup table, obtain Morton codes of N coordinate differences;

The Morton codes of the reference point and the Morton codes of N coordinate differences are calculated to obtain the Morton codes of the N neighbor nodes to be searched.

It should be noted that the Morton codes of N coordinate difference values can be obtained from the preset lookup table. After the Morton code of the reference point is determined, the Morton code of the reference point and the N coordinate difference values can be determined. The Morton code is calculated to obtain the Morton codes of the N neighbor nodes to be searched.

Specifically, the first stored Morton code is obtained from the preset lookup table, and then the Morton code is added to the Morton code of the reference point to obtain the first neighbor to be searched. By analogy, the Morton codes of the N neighbor nodes to be searched can be obtained.

In this way, after obtaining the Morton codes of the N neighbor nodes to be searched, the subsequent steps of judging whether the neighbor nodes to be searched exist can be performed.

S402: If there is a coplanar neighbor node that is coplanar with the current node and has been encoded among the N neighbor nodes to be searched, acquire a first attribute reconstruction value of the coplanar neighbor node.

S403: If there is a collinear neighbor node that is collinear with the current node and has been coded among the N neighbor nodes to be searched, obtain a second attribute reconstruction value of the collinear neighbor node.

S404: Determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.

It should be noted that due to the non-uniform distribution of point clouds, the Morton codes of the neighbor nodes to be searched that are calculated at this time do not necessarily exist; therefore, it is necessary to further determine whether these points exist.

In a possible implementation manner, after determining the Morton codes of the N neighbor nodes to be searched, the method may further include:

Traverse the preset point cloud sequence, and determine whether the preset point cloud sequence has a Morton code of the first neighbor node to be searched;

If the judgment result is yes, it is determined that the first neighbor node to be searched exists in the N neighbor nodes to be searched;

If the judgment result is no, it is determined that the first neighbor node to be searched does not exist in the N neighbor nodes to be searched;

Wherein, the first neighbor node to be searched is any neighbor node to be searched among the N neighbor nodes to be searched.

It should be noted that what is stored in the preset point cloud sequence is the reconstructed point cloud, that is, the points in the preset point cloud sequence are all encoded nodes. For the preset point cloud sequence, in a specific embodiment, the method may further include:

determining a Morton code of at least one point in the point cloud comprising the current node;

Arrange the Morton codes in ascending order to obtain a Morton code sequence;

A Morton code smaller than the Morton code of the current node is intercepted from the Morton code sequence, and the preset point cloud sequence is obtained according to the intercepted Morton code.

It should be noted that the point cloud includes the current node, calculate the Morton code of at least one point in the point cloud, and then arrange the Morton codes in ascending order to obtain the Morton code sequence; The Morton code smaller than the Morton code of the current node is intercepted from the code sequence, so that a preset point cloud sequence can be obtained according to the intercepted Morton code, and each node in the preset point cloud sequence is encoded. node. It should also be noted that, in this embodiment of the present application, the Morton codes can also be arranged in descending order to obtain a Morton code sequence, but the Morton code sequence smaller than that of the current node is still intercepted from the Morton code sequence. Morton code to get preset point cloud sequence.

In this way, by traversing the preset point cloud sequence, when the Morton code of the first neighbor node to be searched exists in the preset point cloud sequence, it can be determined that the first neighbor node to be searched exists in the N neighbor nodes to be searched. node, that is, the first neighbor node to be searched is an encoded node; when there is no Morton code of the first neighbor node to be searched in the preset point cloud sequence, the N neighbor nodes to be searched can be determined at this time. The first neighbor node to be searched does not exist in , that is, the first neighbor node to be searched is a node that has not been coded, that is, it cannot be used as a reference neighbor node for attribute prediction.

In another possible implementation manner, after determining the Morton codes of the N neighbor nodes to be searched, the method may further include:

Determine whether the Morton code of the first neighbor node to be searched is smaller than the Morton code of the current node;

It should be noted that when searching for neighbor nodes, the neighbor nodes used for prediction need to be nodes that have been traversed, and the traversal order is in the order of Morton code from small to large. Therefore, only neighbor nodes whose Morton code is smaller than that of the current node can be used for attribute prediction. If the Morton code of the found neighbor node is larger than that of the current node, it cannot be used as a reference neighbor node. In other words, only when the Morton code of the first neighbor node to be searched is smaller than the Morton code of the current node, it can be determined that the first neighbor node to be searched exists among the N neighbor nodes to be searched, that is, the first neighbor node to be searched. The neighbor node to be searched is an encoded node; and when the Morton code of the first neighbor node to be searched is greater than the Morton code of the current node, it can be determined that the N neighbor nodes to be searched do not exist in the first neighbor node. Once the neighbor node to be searched, that is, the first neighbor node to be searched is a node that has not been coded, that is, it cannot be used as a reference neighbor node for attribute prediction.

Further, in some embodiments, when the first neighbor node to be searched exists in the N neighbor nodes to be searched, the method may further include:

If the first neighbor node to be searched is a coplanar neighbor node of the current node, it is determined that there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched;

If the first neighbor node to be searched is a collinear neighbor node of the current node, it is determined that there is a collinear neighbor node that is collinear with the current node and has been coded among the N neighbor nodes to be searched.

That is to say, if there is a first neighbor node to be searched among the N neighbor nodes to be searched, it is also necessary to judge whether the first neighbor node is a coplanar neighbor node or a collinear neighbor node at this time. Specifically, if the first neighbor node to be searched is a coplanar neighbor node of the current node, it can be determined that there are coplanar neighbor nodes that are coplanar with the current node and have been encoded in the N neighbor nodes to be searched; If the neighbor node is the collinear neighbor node of the current node, it can be determined that there are collinear neighbor nodes that are collinear with the current node and have been encoded among the N neighbor nodes to be searched.

In this embodiment of the present application, the closer the distance, the stronger the spatial correlation. The distance between the current node and the coplanar neighbor node is 1, and the distance between the current node and the collinear neighbor node is 1

Therefore, relatively speaking, coplanar neighbor nodes have stronger spatial correlation, and in this embodiment of the present application, it is possible to preferentially determine whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded among the N neighbor nodes to be searched.

It should be noted that, in some embodiments, the method may further include: setting the number of reference neighbor nodes to K, where K is an integer greater than 0 and less than or equal to N.

That is, in this embodiment of the present application, the number of reference neighbor nodes may be preset. In this way, among the N neighbor nodes to be searched, when K existing neighbor nodes are searched, the search can be stopped at this time, and the K neighbor nodes obtained by the search can be directly used for attribute prediction, thereby reducing the search complexity.

Wherein, the value of K may be 3, may also be 6, and may even be 8 or other values. In the embodiment of the present application, the value of K is specifically set according to the actual situation, and no limitation is made here.

It should also be noted that, in some embodiments, the method may further include: writing the value of K into the code stream.

In this way, since the encoder writes the value of K into the code stream, the subsequent decoder can directly obtain the number of reference neighbor nodes (the value of K) by parsing the code stream, so that the predictions used for attribute prediction in the encoder and decoder are Let the number of reference neighbor nodes be the same.

In a possible implementation manner, when it is found that the existing coplanar neighbor nodes in the N neighbor nodes to be searched already satisfy the requirement of K, at this time, only the K coplanar neighbor nodes can be used for attribute prediction. In some embodiments, the method may also include:

Judging whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched;

If there are K coplanar neighbor nodes that are coplanar with the current node and have been encoded in the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the K coplanar neighbor nodes;

Correspondingly, for S404, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value may include:

Perform mean calculation on the first attribute reconstruction values of the K coplanar neighbor nodes to determine the attribute prediction value of the current node.

It should be noted that, in the N neighbor nodes to be searched, it is searched whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded. When K existing coplanar neighbor nodes are found, the search can be stopped at this time. ; and then obtain the first attribute reconstruction value of the K coplanar neighbor nodes, and calculate the mean value of the first attribute reconstruction value of the K coplanar neighbor nodes to obtain the attribute prediction value of the current node.

In another possible implementation, when it is found that the existing collinear neighbor nodes in the N neighbor nodes to be searched already satisfy the requirement of K, at this time, only the K collinear neighbor nodes can be used for attribute prediction. In some embodiments, the method may further include:

Judging whether there is a collinear neighbor node that is collinear with the current node and has been coded in the N neighbor nodes to be searched;

If there are K collinear neighbor nodes that are collinear with the current node and have been coded in the N neighbor nodes to be searched, acquiring the second attribute reconstruction value of the K collinear neighbor nodes;

Perform mean calculation on the second attribute reconstruction values of the K collinear neighbor nodes to determine the attribute prediction value of the current node.

It should be noted that, in the N neighbor nodes to be searched, it is searched whether there is a collinear neighbor node that is collinear with the current node and has been encoded. When K existing collinear neighbor nodes are searched, the search can be stopped at this time. ; and then obtain the second attribute reconstruction value of the K collinear neighbor nodes, and calculate the mean value of the second attribute reconstruction value of the K collinear neighbor nodes to obtain the attribute prediction value of the current node.

In another possible implementation manner, considering that the coplanar neighbor nodes have stronger spatial correlation, it is necessary to first search whether there are coplanar neighbor nodes, and then search whether there are collinear neighbor nodes.

When no existing coplanar neighbor nodes are found in the N neighbor nodes to be searched, but the existing collinear neighbor nodes are found to meet the requirements of K, at this time, only the K collinear neighbor nodes can be used for attribute prediction. . In some embodiments, the method may also include:

If there is no coplanar neighbor node that is coplanar with the current node and has been coded in the N neighbor nodes to be searched, it is determined whether there is a coplanar neighbor node that is collinear with the current node and has been coded in the N neighbor nodes to be searched. encoded collinear neighbor nodes;

It should be noted that, firstly, in the N neighbor nodes to be searched, it is searched whether there is a coplanar neighbor node that is coplanar with the current node and has been coded. Search the neighbor nodes to see if there is a collinear neighbor node that is collinear with the current node and has been coded. When K existing collinear neighbor nodes are searched, the search can be stopped at this time, and then the information of the K collinear neighbor nodes can be obtained. The second attribute reconstruction value, and by performing mean calculation on the second attribute reconstruction value of the K collinear neighbor nodes, the attribute prediction value of the current node can be obtained.

Here, it should be noted that if the N neighbor nodes to be searched are searched for whether there are collinear neighbor nodes that are collinear with the current node and have been coded, only k existing collinear neighbor nodes can be searched, and k is an integer greater than 0 and less than K. At this time, the predicted value of the attribute of the current node can be obtained by performing mean calculation on the reconstructed value of the second attribute of the k collinear neighbor nodes.

In yet another possible implementation, when M existing coplanar neighbor nodes are found in the N neighbor nodes to be searched, but M is less than K, it is necessary to continue to search for L existing colinear neighbor nodes, and then Use these M coplanar neighbor nodes and L collinear neighbor nodes to jointly perform attribute prediction. In some embodiments, the method may also include:

If there are M coplanar neighbor nodes that are coplanar with the current node and have been encoded among the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the M coplanar neighbor nodes, where M is greater than 0 and an integer less than K;

If there are L collinear neighbor nodes that are collinear with the current node and have been coded in the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the L collinear neighbor nodes; wherein, L is an integer greater than 0 and less than or equal to (K-M);

A weighted calculation is performed on the M first attribute reconstruction values and the L second attribute reconstruction values to determine the attribute prediction value of the current node.

It should be noted that, firstly, among the N neighbor nodes to be searched, it is searched whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded. If M existing coplanar neighbor nodes are found, but M is less than K, then It is necessary to continue to search the N neighbor nodes to be searched for whether there are collinear neighbor nodes that are collinear with the current node and have been coded. When (K-M) existing coplanar neighbor nodes are found, the search can be stopped at this time, and then Obtain the first attribute reconstruction values of the M coplanar neighbor nodes and the second attribute reconstruction values of the (K-M) collinear neighbor nodes, and obtain the M first attribute reconstruction values and the (K-M) second attribute reconstruction values. The attribute reconstruction value is weighted to calculate the attribute prediction value of the current node.

Here, it should be noted that if the N neighbor nodes to be searched are searched for whether there are collinear neighbor nodes that are collinear with the current node and have been coded, only L existing collinear neighbor nodes can be searched, and L is an integer greater than 0 and less than or equal to (K-M), at this time, the attribute prediction value of the current node can be obtained by performing weighted calculation on the M first attribute reconstruction values and the L second attribute reconstruction values.

Further, in some embodiments, the performing weighted calculation on the M first attribute reconstruction values and the L second attribute reconstruction values to determine the attribute prediction value of the current node may include:

determining the first weight values corresponding to the M first attribute reconstruction values and the second weight values corresponding to the L second attribute reconstruction values;

Perform a weighted average calculation according to the M first attribute reconstruction values and the corresponding first weight values and the L second attribute reconstruction values and the corresponding second weight values to obtain the attribute of the current node Predictive value.

It should be noted that, in consideration of the strength of spatial correlation, in this embodiment of the present application, the first weight value (that is, the weight of the coplanar neighbor node) may be set to 2, and the second weight value (that is, the weight of the collinear neighbor node) Set to 1.

It should also be noted that it is assumed that the reconstruction value of the first attribute is

represents, the first weight value is represented by w _1i , i=1,2,...,M; the second attribute reconstruction value is represented by

represents, the second weight value is represented by w _2j , j=1,2,...,L; the attribute prediction value of the current node is represented by A, then its calculation formula is as follows:

In addition, when neither an existing coplanar neighbor node nor an existing coplanar neighbor node is found in the N neighbor nodes to be searched, at this time, in some embodiments, the method may further include:

If there is no coplanar neighbor node that is coplanar with the current node and has been coded and there is no colinear neighbor node that is colinear with the current node and has been coded in the N neighbor nodes to be searched, then In the Morton code sequence, the previous Morton code of the Morton code located at the current node is determined, and the attribute reconstruction value of the point corresponding to the previous Morton code is determined as the attribute prediction value of the current node.

That is to say, when neither the existing coplanar neighbor nodes nor the existing coplanar neighbor nodes are found in the N neighbor nodes to be searched, the Morton code located at the current node in the Morton code sequence can be directly used at this time. The attribute reconstruction value of the previous point of the code is used as the attribute prediction value of the current node. In other words, for the points in the point cloud, except for the first point that does not need attribute prediction, all other points need attribute prediction.

This embodiment provides a prediction method, by determining N neighbor nodes to be searched of the current node; if there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched, then Obtain the first attribute reconstruction value of the coplanar neighbor node; if there is a collinear neighbor node that is collinear with the current node and has been encoded in the N neighbor nodes to be searched, then obtain the collinear neighbor node. second attribute reconstruction value; determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value. In this way, when the attribute prediction is performed according to the geometric space relationship between the current node and the neighbor nodes, due to the expansion of the neighbor search range (for example, from 6 neighbor nodes to be searched in the related art to N neighbor nodes to be searched), it is possible to Make full use of the spatial correlation of point clouds to make the prediction residual smaller, which can reduce the code rate and improve the coding efficiency.

In another embodiment of the present application, referring to FIG. 5 , it shows a schematic flowchart of another prediction method provided by the embodiment of the present application. As shown in Figure 5, the method may include:

S501: Determine N neighbor nodes to be searched for the current node.

Here, compared with the related art, the embodiment of the present application expands the neighbor search range. Among them, the value of N can be set to 12.

As shown in FIG. 6A , it shows a schematic diagram of a neighbor search range provided by the related art. As shown in FIG. 6B , it shows a schematic diagram of a neighbor search range provided by an embodiment of the present application. In Fig. 6A, the block filled with gray is the current node, and its neighbor search range includes 6 neighbor nodes to be searched. In FIG. 6B , the block filled with gray is the current node, and its neighbor search range is extended by another 6 neighbor nodes to be searched in addition to the 6 neighbor nodes to be searched as shown in FIG. 6A . That is to say, the neighbor search range in this embodiment of the present application may include 12 neighbor nodes to be searched.

S502: Set the number of reference neighbor nodes to K, where K is an integer greater than 0 and less than or equal to N.

It should be noted that the encoder can preset the number of reference neighbor nodes. In this way, when K existing neighbor nodes are found in the N neighbor nodes to be searched, the search can be stopped at this time, which reduces the search complexity. Subsequently, the K neighbor nodes can be used as a reference to predict the attributes of the current node.

S503: In the case where M coplanar neighbor nodes that are coplanar with the current node and have been encoded are found in the N neighbor nodes to be searched, obtain first attribute reconstruction of the M coplanar neighbor nodes value.

S504: Determine whether M is equal to K.

S505: When M is equal to K, perform mean calculation on the first attribute reconstruction values of the M coplanar neighbor nodes, and determine the attribute prediction value of the current node.

S506: When M is less than K, in the case where there are L collinear neighbor nodes that are collinear with the current node and have been coded in the N neighbor nodes to be searched, obtain the L collinear neighbors The second attribute of the node reconstructs the value.

S507: Perform weighted calculation on the M first attribute reconstruction values and the L second attribute reconstruction values to determine the attribute prediction value of the current node.

Here, M is an integer greater than 0, and L is an integer greater than 0 and less than or equal to (K-M).

It should be noted that, considering the stronger spatial correlation of coplanar neighbor nodes, it is necessary to first search for the existence of coplanar neighbor nodes, and then to search for the existence of collinear neighbor nodes.

It should also be noted that, after obtaining the first attribute reconstruction values of the M coplanar neighbor nodes, for S504, if M is equal to K at this time, then execute S505; if M is less than K, execute S506-S507.

In addition, when M is equal to 0, that is, no existing coplanar neighbor nodes are found in the N neighbor nodes to be searched, it is necessary to continue to search for the existence of the collinear neighbor nodes in the N neighbor nodes to be searched; if At this time, L existing collinear neighbor nodes are searched, and the mean value is calculated by the second attribute reconstruction value of the L collinear neighbor nodes, and the attribute prediction value of the current node can be obtained.

In addition, when M is equal to 0 and L is equal to 0, that is, there is no coplanar neighbor node that exists in the N neighbor nodes to be searched, and no colinear neighbor node is found. At this time, you can directly use the The attribute reconstruction value of the previous point of the Morton code located at the current node in the Morton code sequence is used as the attribute prediction value of the current node.

In short, the embodiment of the present application proposes an attribute prediction method for expanding the neighbor search range. First, operate the Morton code (cur_pos) of the current node to locate the reference point Morton code (base_pos); then calculate the coordinate difference between the reference point and the N adjacent nodes to be searched for the current node; Morton coding is performed on these coordinate differences, and a preset look-up table is established; and then Morton codes are added according to the preset look-up table to determine the Morton codes of the N adjacent nodes to be searched.

Due to the non-uniform distribution of point clouds, the calculated Morton codes of neighboring nodes do not necessarily exist. At this time, it is also necessary to traverse the preset point cloud sequence to determine whether these points exist. Here, the preset point cloud sequence is a reconstructed point cloud, and the points in the preset point cloud sequence are all coded nodes.

Exemplarily, as shown in FIG. 6A , the block filled with gray is the current node, and its neighbor search range includes 6 neighbor nodes to be searched. As shown in Figure 6B, the block filled with gray is the current node, and its neighbor search range is extended by another 6 neighbor nodes to be searched in addition to the 6 neighbor nodes to be searched as shown in Figure 6A (indicated by bold lines). ). That is to say, the neighbor search range in this embodiment of the present application may include 12 neighbor nodes to be searched. Specifically, when searching for neighbor nodes in the embodiment of the present application, the neighbor nodes used for prediction are points that have been traversed, and the traversal order is in the order of Morton codes from small to large. Therefore, only the point whose Morton code is smaller than the current point can be used for prediction. If the Morton code of the found neighbor node is larger than that of the current node, it cannot be used as a reference neighbor node.

In this way, since the embodiment of the present application considers that the closer the distance is, the stronger the correlation is, then a coplanar neighbor node with a distance of 1 may be searched first, and if it exists, the average value of these coplanar neighbor nodes is used as the predicted value; or, if If there is no coplanar neighbor node at all, then you can continue to search for a distance of

The average value of these collinear neighbor nodes is used as the predicted value; or, if there are coplanar neighbor nodes but the preset number of reference neighbor nodes is not met, then the search can also be continued. The search distance is

The collinear neighbor nodes of , use the weighted average of these coplanar neighbor nodes and the collinear neighbor nodes as the predicted value; if none of the above neighbor nodes exist, then the previous point of the Morton code sequence of the current node can be used at this time. The attribute reconstruction value of is directly used as the attribute prediction value.

In this way, the embodiment of the present application expands the neighbor search range, and can make full use of the spatial correlation of the point cloud, so that the prediction residual is smaller, which is more suitable for subsequent quantization and entropy coding. Here, on the premise that the technical solutions proposed in the embodiments of the present application do not substantially increase the complexity of encoding and decoding, using the technical solutions has the following beneficial effects:

The technical solutions of the embodiments of the present application can reduce the size of the encoded code stream without changing the attribute PSNR. Here, PSNR is an objective criterion for image evaluation. The larger the PSNR, the better the image quality. On the AVS platform, after testing the sequence with part of the attribute information as color under the condition of C2 (geometric lossless, attribute lossy), the test results are shown in Table 1 below.

Table 1

As can be seen from Table 1 above, the BD-rates of all test sequences are negative, and when the BD-rate is negative, the performance is improved. On this basis, the greater the absolute value of BD-rate, the greater the performance gain.

This embodiment provides a prediction method. The specific implementation of the foregoing embodiment is described in detail through the foregoing embodiment. It can be seen from this that the present application expands the neighbor search range when performing attribute prediction according to the geometric space relationship, and fully The spatial correlation of adjacent points is exploited. Compared with the related art, the technical solution of the present application utilizes the spatial correlation of the point cloud to a greater extent, and under the condition that the PSNR performance is maintained or even slightly better, the output binary code stream can be reduced, so that the code rate can be reduced, thereby improving the performance. coding efficiency.

In another embodiment of the present application, the prediction method provided by the embodiment of the present application is applied to a video decoding device, that is, a point cloud decoder (or referred to as a PCRM decoder), and the embodiment of the present application may also be referred to as a decoder for short. The functions implemented by the method can be implemented by calling a computer program by the second processor in the decoder. Of course, the computer program can be stored in the second memory. It can be seen that the decoder includes at least the second processor and the second memory.

Referring to FIG. 7 , it shows a schematic flowchart of another prediction method provided by an embodiment of the present application. As shown in Figure 7, the method may include:

S701: Determine N neighbor nodes to be searched for the current node.

It should be noted that, in a point cloud, a point can be all points in the point cloud, or some points in the point cloud, and these points are relatively concentrated in space. Here, the current node may also be referred to as the current point or the current block, and specifically refers to the point or block currently to be decoded in the point cloud.

It should also be noted that, based on the framework of the PCRM decoder shown in FIG. 1B , the method in the embodiment of the present application is mainly applied to the “attribute reconstruction” part, and is optimized for the attribute prediction based on the neighbor search range in the current related art, through Expand the neighbor search range to improve the utilization of spatial correlation.

Perform spatial division on the point cloud to obtain at least one point;

Three neighbor nodes that must have been decoded in the six neighbor nodes that are coplanar with the current node are determined as three neighbor nodes to be searched, and the twelve neighbor nodes that are collinear with the current node must have been decoded. Three neighbor nodes are determined as three neighbor nodes to be searched;

Based on the geometric space relationship between the current node and the neighbor nodes, n neighbor nodes that may have been decoded among the remaining nine neighbor nodes collinear with the current node are determined as the n neighbor nodes to be searched, where n is greater than an integer of 0 and less than 9;

It should be noted that, among the 26 neighbor nodes, in the current related technology, only six neighbor nodes that can be determined to have been decoded are formed into a neighbor search range. Specifically, as shown in FIG. 3A, among the six neighbor nodes coplanar with the current node, there are only three neighbor nodes coplanar with the left side, front side, and lower side of the current node (respectively with 3, 5, 6 padding) is able to determine that it must have been decoded. As shown in Figure 3B, among the twelve neighbor nodes that are collinear with the current node, there are only three neighbor nodes that are collinear with the left, front, and bottom directions of the current node (filled with 1, 2, and 4, respectively) It can be determined that it must have been decoded.

In the embodiment of the present application, in addition to determining the six neighbor nodes that have been decoded, the embodiment of the present application expands the neighbor search range. That is to say, based on the geometric space relationship between the current node and the neighbor nodes, the n neighbor nodes that may have been decoded among the remaining nine neighbor nodes collinear with the current node are determined as the n neighbor nodes to be searched, so that the expansion The subsequent neighbor search range is (6+n) neighbor nodes to be searched. Here, n is an integer greater than 0 and less than 9, that is, N is an integer greater than 6.

It should also be noted that in the process of expanding the neighbor search range, since the remaining three neighbor nodes that are coplanar with the current node can be sure that they have not been decoded, they are not considered; Neighbor nodes can also determine that they must not have been decoded, so they are not considered; and three of the remaining nine neighbor nodes that are collinear with the current node can determine that they must not have been decoded. Therefore, in this application In a specific example of the embodiment, n is an integer greater than 0 and less than or equal to 6.

Further, in some embodiments, the method may further include: parsing the code stream to obtain the value of N.

In this way, since the encoder writes the value of N into the code stream, the decoder can directly parse the code stream to obtain the number of neighbor nodes to be searched (the value of N), thereby making the neighbor search range of the encoder and decoder be consistent.

Determine the reference point of the current node;

calculating the Morton code of the current node;

In the embodiment of the present application, the Morton code corresponding to each point in the point cloud can be obtained by using the coordinate information of the node (or simply referred to as "point"). Here, Morton code can convert multi-dimensional data into one-dimensional data encoding, and Morton code can also be called z-order code, because its decoding order is also in the spatial z order. In this way, after the reference point is obtained, the coordinate difference (offset) between the reference point and the N neighbor nodes to be searched can be calculated, and the N coordinate difference values can be converted into N Morton codes, and then the N Frames are stored in a preset lookup table.

Further, determining the Morton codes of the N neighbor nodes to be searched according to the preset lookup table may include:

Determine the Morton code of the reference point;

From the preset lookup table, obtain Morton codes of N coordinate differences;

The Morton code of the reference point and the Morton code of N coordinate differences are calculated to obtain the Morton codes of the N neighbor nodes to be searched.

It should be noted that the Morton codes of N coordinate differences can be obtained from the preset lookup table. After the Morton codes of the reference points are determined, the Morton codes of the reference points and the N coordinate differences can be analyzed. The frame code is calculated to obtain the Morton code of the N neighbor nodes to be searched.

S702: If there is a coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, obtain a first attribute reconstruction value of the coplanar neighbor node.

S703: If there is a collinear neighbor node that is collinear with the current node and has been decoded among the N neighbor nodes to be searched, obtain a second attribute reconstruction value of the collinear neighbor node.

S704: Determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.

It should be noted that what is stored in the preset point cloud sequence is the reconstructed point cloud, that is, the points in the preset point cloud sequence are all decoded nodes. For the preset point cloud sequence, in a specific embodiment, the method may further include:

Arrange the Morton codes in ascending order to obtain a Morton code sequence;

It should be noted that when searching for neighbor nodes, the neighbor nodes used for prediction need to be nodes that have been traversed, and the traversal order is in the order of Morton code from small to large. Therefore, only neighbor nodes whose Morton code is smaller than that of the current node can be used for attribute prediction. If the Morton code of the found neighbor node is larger than that of the current node, it cannot be used as a reference neighbor node. In other words, only when the Morton code of the first neighbor node to be searched is smaller than the Morton code of the current node, it can be determined that the first neighbor node to be searched exists among the N neighbor nodes to be searched, that is, the first neighbor node to be searched. The neighbor node to be searched is a decoded node; and when the Morton code of the first neighbor node to be searched is greater than the Morton code of the current node, it can be determined that the N neighbor nodes to be searched do not exist in the first neighbor node. Once the neighbor node to be searched, that is, the first neighbor node to be searched is a node that has not been decoded, that is, it cannot be used as a reference neighbor node for attribute prediction.

If the first neighbor node to be searched is a coplanar neighbor node of the current node, it is determined that there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched;

If the first neighbor node to be searched is a collinear neighbor node of the current node, it is determined that there is a collinear neighbor node that is collinear with the current node and has been decoded among the N neighbor nodes to be searched.

That is to say, if there is a first neighbor node to be searched among the N neighbor nodes to be searched, it is also necessary to judge whether the first neighbor node is a coplanar neighbor node or a collinear neighbor node at this time. Specifically, if the first neighbor node to be searched is a coplanar neighbor node of the current node, it can be determined that there are coplanar neighbor nodes that are coplanar with the current node and have been decoded in the N neighbor nodes to be searched; If the neighbor node is a collinear neighbor node of the current node, it can be determined that there is a collinear neighbor node that is collinear with the current node and has been decoded among the N neighbor nodes to be searched.

Therefore, relatively speaking, coplanar neighbor nodes have stronger spatial correlation, and in this embodiment of the present application, it is possible to preferentially determine whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched.

It should be noted that, in this embodiment of the present application, the number of reference neighbor nodes may also be preset. In some embodiments, the method may further include: setting the number of reference neighbor nodes to K, where K is an integer greater than 0 and less than or equal to N.

It should also be noted that, in this embodiment of the present application, the value of K may also be written into the code stream on the encoder side. In some embodiments, the method may further include: parsing the code stream to obtain the value of K; where K represents the number of reference neighbor nodes.

In this way, among the N neighbor nodes to be searched, when K existing neighbor nodes are searched, the search can be stopped at this time, and the K neighbor nodes obtained by the search can be directly used for attribute prediction, thereby reducing the search complexity.

Determine whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched;

If there are K coplanar neighbor nodes that are coplanar with the current node and have been decoded in the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the K coplanar neighbor nodes;

Correspondingly, for S704, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value may include:

It should be noted that the N neighbor nodes to be searched are searched for whether there are coplanar neighbor nodes that are coplanar with the current node and have been decoded. When K existing coplanar neighbor nodes are found, the search can be stopped at this time. ; and then obtain the first attribute reconstruction value of the K coplanar neighbor nodes, and calculate the mean value of the first attribute reconstruction value of the K coplanar neighbor nodes to obtain the attribute prediction value of the current node.

Determine whether there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched;

If there are K collinear neighbor nodes that are collinear with the current node and have been decoded in the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the K collinear neighbor nodes;

It should be noted that the N neighbor nodes to be searched are searched for whether there are collinear neighbor nodes that are collinear with the current node and have been decoded. When K existing collinear neighbor nodes are searched, the search can be stopped at this time. ; and then obtain the second attribute reconstruction value of the K collinear neighbor nodes, and calculate the mean value of the second attribute reconstruction value of the K collinear neighbor nodes to obtain the attribute prediction value of the current node.

If there is no coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, it is judged whether there is a coplanar neighbor node that is collinear with the current node and has been decoded among the N neighbor nodes to be searched. Decoded collinear neighbor nodes;

It should be noted that, firstly, in the N neighbor nodes to be searched, it is searched whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded. Search the neighbor nodes to see if there are collinear neighbor nodes that are collinear with the current node and have been decoded. When K existing collinear neighbor nodes are searched, the search can be stopped at this time, and then the K collinear neighbor nodes can be obtained. The second attribute reconstruction value, and by performing mean calculation on the second attribute reconstruction value of the K collinear neighbor nodes, the attribute prediction value of the current node can be obtained.

Here, it should be noted that if the N neighbor nodes to be searched are searched for whether there are collinear neighbor nodes that are collinear with the current node and have been decoded, only k existing collinear neighbor nodes can be searched, and k is an integer greater than 0 and less than K. At this time, the predicted value of the attribute of the current node can be obtained by performing mean calculation on the reconstructed value of the second attribute of the k collinear neighbor nodes.

If there are M coplanar neighbor nodes that are coplanar with the current node and have been decoded among the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the M coplanar neighbor nodes, where M is greater than 0 and an integer less than K;

If there are L collinear neighbor nodes that are collinear with the current node and have been decoded among the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the L collinear neighbor nodes; wherein, L is an integer greater than 0 and less than or equal to (K-M);

It should be noted that, firstly, the N neighbor nodes to be searched are searched for whether there are coplanar neighbor nodes that are coplanar with the current node and have been decoded. If M existing coplanar neighbor nodes are found, but M is less than K, then It is necessary to continue to search the N neighbor nodes to be searched for whether there are collinear neighbor nodes that are collinear with the current node and have been decoded. When (K-M) existing coplanar neighbor nodes are found, the search can be stopped at this time, and then Obtain the first attribute reconstruction values of the M coplanar neighbor nodes and the second attribute reconstruction values of the (K-M) collinear neighbor nodes, and obtain the M first attribute reconstruction values and the (K-M) second attribute reconstruction values. The attribute reconstruction value is weighted to calculate the attribute prediction value of the current node.

Here, it should be noted that if the N neighbor nodes to be searched are searched for whether there are collinear neighbor nodes that are collinear with the current node and have been decoded, only L existing collinear neighbor nodes can be searched, and L is an integer greater than 0 and less than or equal to (K-M), at this time, the attribute prediction value of the current node can be obtained by performing weighted calculation on the M first attribute reconstruction values and the L second attribute reconstruction values.

It should be noted that, in consideration of the strength of spatial correlation, in this embodiment of the present application, the first weight value (that is, the weight of the coplanar neighbor node) may be set to 2, and the second weight value (that is, the weight of the collinear neighbor node) Set to 1. Specifically, weighted calculation can be performed according to formula (2) to obtain the attribute prediction value of the current node.

If there is no coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, and there is no colinear neighbor node that is collinear with the current node and has been decoded, then In the Morton code sequence, the previous Morton code of the Morton code located at the current node is determined, and the attribute reconstruction value of the point corresponding to the previous Morton code is determined as the attribute prediction value of the current node.

This embodiment provides a prediction method, by determining N neighbor nodes to be searched of the current node; if there is a coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, then Obtain the first attribute reconstruction value of the coplanar neighbor node; if there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched, then obtain the collinear neighbor node. second attribute reconstruction value; determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value. In this way, when the attribute prediction is performed according to the geometric space relationship between the current node and the neighbor nodes, due to the expansion of the neighbor search range (for example, from 6 neighbor nodes to be searched in the related art to N neighbor nodes to be searched), it is possible to Making full use of the spatial correlation of the point cloud makes the prediction residual smaller, which can reduce the bit rate and improve the decoding efficiency.

In yet another embodiment of the present application, based on the same inventive concept as the foregoing embodiments, see FIG. 8 , which shows a schematic structural diagram of an encoder 80 provided by an embodiment of the present application. As shown in FIG. 8 , the encoder 80 may include: a first determination unit 801, a first judgment unit 802 and a first prediction unit 803; wherein,

The first determining unit 801 is configured to determine the N neighbor nodes to be searched for the current node;

The first judgment unit 802 is configured to obtain a first attribute reconstruction value of the coplanar neighbor node if there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched;

The first judging unit 802 is further configured to obtain a second attribute reconstruction value of the collinear neighbor node if there is a collinear neighbor node that is collinear with the current node and has been coded in the N neighbor nodes to be searched ;

The first prediction unit 803 is configured to determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.

In some embodiments, referring to FIG. 8 , the encoder 80 may further include a first dividing unit 804 configured to perform spatial division on the point cloud to obtain at least one point;

The first determining unit 801 is further configured to, based on the geometric position of the current node, determine the neighbor information of the current node from the at least one point; and determine the N neighbor information from the neighbor information of the current node Neighbor node to be searched.

In some embodiments, the neighbor information of the current node includes: six neighbor nodes coplanar with the current node, twelve neighbor nodes collinear with the current node, and co-located with the current node Eight neighbor nodes.

In some embodiments, the first determining unit 801 is specifically configured to determine three neighbor nodes that must have been encoded among the six neighbor nodes coplanar with the current node as the three neighbor nodes to be searched, Among the twelve neighbor nodes collinear with the current node, three neighbor nodes that must have been encoded are determined as three neighbor nodes to be searched; and based on the geometric space relationship between the current node and the neighbor nodes, the The n neighbor nodes that may have been encoded in the remaining nine neighbor nodes that are collinear are determined as n neighbor nodes to be searched, n is an integer greater than 0 and less than 9; and the obtained (6+n) neighbor nodes to be searched It is determined as the N neighbor nodes to be searched.

In some embodiments, N is an integer greater than 6.

In some embodiments, referring to FIG. 8 , the encoder 80 may further include a first computing unit 805;

The first determining unit 801 is further configured to determine the reference point of the current node;

The first calculation unit 805 is configured to perform difference calculation on the coordinate information between the reference point and the N neighbor nodes to be searched to obtain N coordinate difference values; and a method for calculating the N coordinate difference values. frame code, and the Morton code of the N coordinate differences is stored in a preset look-up table;

The first determining unit 801 is further configured to determine Morton codes of the N neighbor nodes to be searched according to the preset lookup table.

In some embodiments, the first calculation unit 805 is further configured to calculate the Morton code of the current node;

The first determining unit 801 is further configured to, according to the Morton code of the current node, determine the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node; and determine the neighbor node corresponding to the smallest Morton code Determined as the reference point.

In some embodiments, the first determining unit 801 is specifically configured to determine the coordinate information of the current node according to the Morton code of the current node; and determine the neighbor information of the current node based on the coordinate information of the current node and selecting the smallest Morton code from the Morton codes of the at least one neighbor node to obtain the neighbor node corresponding to the smallest Morton code.

In some embodiments, the first determining unit 801 is further configured to determine the Morton code of the reference point;

The first calculation unit 805 is further configured to obtain the Morton code of the N coordinate difference values from the preset lookup table; and the Morton code of the reference point and the N coordinate difference values. The Morton code is calculated to obtain the Morton codes of the N neighbor nodes to be searched.

In some embodiments, the first determination unit 802 is further configured to traverse a preset point cloud sequence, and determine whether the preset point cloud sequence has a Morton code of the first neighbor node to be searched;

The first determining unit 801 is further configured to determine that the first neighbor node to be searched exists in the N neighbor nodes to be searched if the determination result is yes; and to determine the N neighbor nodes to be searched if the determination result is no The first neighbor node to be searched does not exist in the neighbor nodes; wherein, the first neighbor node to be searched is any one of the N neighbor nodes to be searched.

In some embodiments, referring to FIG. 8 , the encoder 80 may further include a first sorting unit 806;

The first determining unit 801 is further configured to determine the Morton code of at least one point in the point cloud including the current node;

The first sorting unit 806 is configured to arrange the Morton codes in ascending order to obtain a Morton code sequence; and truncate a Morton code smaller than the current node's Morton code from the Morton code sequence Morton code, and obtain the preset point cloud sequence according to the intercepted Morton code.

In some embodiments, the first judgment unit 802 is further configured to judge whether the Morton code of the first neighbor node to be searched is smaller than the Morton code of the current node;

In some embodiments, the first determining unit 801 is further configured to, when there is a first neighbor node to be searched in the N neighbor nodes to be searched, if the first neighbor node to be searched is a coplanar neighbor of the current node node, then determine that there is a coplanar neighbor node that is coplanar with the current node and has been coded in the N neighbor nodes to be searched; if the first neighbor node to be searched is the collinear neighbor node of the current node, then determine Among the N neighbor nodes to be searched, there is a collinear neighbor node that is collinear with the current node and has been coded.

In some embodiments, referring to FIG. 8 , the encoder 80 may further include a writing unit 807 configured to write the value of N into the code stream.

In some embodiments, referring to FIG. 8 , the encoder 80 may further include a first setting unit 808 configured to set the number of reference neighbor nodes to K, where K is an integer greater than 0 and less than or equal to N.

In some embodiments, the first judgment unit 802 is further configured to judge whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded among the N neighbor nodes to be searched;

The first determining unit 801 is further configured to obtain the first number of the K coplanar neighbor nodes if there are K coplanar neighbor nodes that are coplanar with the current node and have been encoded in the N neighbor nodes to be searched. an attribute reconstruction value;

The first prediction unit 803 is specifically configured to perform mean value calculation on the first attribute reconstruction values of the K coplanar neighbor nodes, and determine the attribute prediction value of the current node.

In some embodiments, the first judging unit 802 is further configured to judge whether there is a coplanar neighbor node that is coplanar with the current node and has been coded among the N neighbor nodes to be searched; and if the N neighbor nodes to be searched exist There is no coplanar neighbor node that is coplanar with the current node and has been coded in the search neighbor nodes, then judge whether there is a colinear neighbor node that is colinear with the current node and has been coded in the N neighbor nodes to be searched ;

The first determining unit 801 is further configured to obtain the first number of the K collinear neighbor nodes if there are K collinear neighbor nodes that are collinear with the current node and have been coded in the N neighbor nodes to be searched. Two attribute reconstruction value;

The first prediction unit 803 is specifically configured to perform mean calculation on the second attribute reconstruction values of the K collinear neighbor nodes, and determine the attribute prediction value of the current node.

The first determining unit 801 is further configured to obtain the first number of the M coplanar neighbor nodes if there are M coplanar neighbor nodes that are coplanar with the current node and have been encoded in the N neighbor nodes to be searched. An attribute reconstruction value, M is an integer greater than 0 and less than K;

The first judging unit 802 is further configured to judge whether there is a collinear neighbor node that is collinear with the current node and has been coded in the N neighbor nodes to be searched;

The first determining unit 801 is further configured to obtain the first number of the L collinear neighbor nodes if there are L collinear neighbor nodes that are collinear with the current node and have been coded in the N neighbor nodes to be searched. Two attribute reconstruction values; wherein, L is an integer greater than 0 and less than or equal to (K-M);

The first prediction unit 803 is specifically configured to perform weighted calculation on the M first attribute reconstruction values and the L second attribute reconstruction values, and determine the attribute prediction value of the current node.

In some embodiments, the writing unit 807 is further configured to write the value of K into the code stream.

In some embodiments, the first prediction unit 803 is further configured to: if there is no coplanar neighbor node coplanar with the current node and has been coded among the N neighbor nodes to be searched, and there is no coplanar neighbor node that is coplanar with the current node If there are collinear and coded collinear neighbor nodes, the Morton code before the Morton code located at the current node is determined from the Morton code sequence, and the attribute of the point corresponding to the previous Morton code is determined. The reconstructed value is determined as the attribute prediction value of the current node.

It can be understood that, in the embodiments of the present application, a "unit" may be a part of a circuit, a part of a processor, a part of a program or software, etc., of course, it may also be a module, and it may also be non-modular. Moreover, each component in this embodiment may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of software function modules.

If the integrated unit is implemented in the form of a software functional module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment is essentially or The part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product, the computer software product is stored in a storage medium, and includes several instructions for making a computer device (which can be It is a personal computer, a server, or a network device, etc.) or a processor (processor) that executes all or part of the steps of the method described in this embodiment. The aforementioned storage medium includes: U disk, mobile hard disk, read only memory (Read Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes.

Therefore, an embodiment of the present application provides a computer storage medium, which is applied to the encoder 80, where the computer storage medium stores a computer program, and when the computer program is executed by the first processor, any one of the foregoing embodiments is implemented. Methods.

Based on the composition of the encoder 80 and the computer storage medium described above, see FIG. 9 , which shows a schematic diagram of a specific hardware structure of the encoder 80 provided by the embodiment of the present application. As shown in FIG. 9 , it may include: a first communication interface 901 , a first memory 902 and a first processor 903 ; each component is coupled together through a first bus system 904 . It can be understood that the first bus system 904 is used to realize the connection and communication between these components. In addition to the data bus, the first bus system 904 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, the various buses are labeled as the first bus system 904 in FIG. 9 . in,

The first communication interface 901 is used for receiving and sending signals in the process of sending and receiving information with other external network elements;

a first memory 902 for storing a computer program that can run on the first processor 903;

The first processor 903 is configured to, when running the computer program, execute:

Determine the N neighbor nodes to be searched for the current node;

It can be understood that the first memory 902 in this embodiment of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Wherein, the non-volatile memory may be a read-only memory (Read-Only Memory, ROM), a programmable read-only memory (Programmable ROM, PROM), an erasable programmable read-only memory (Erasable PROM, EPROM), an electrically programmable read-only memory (Erasable PROM, EPROM). Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory. Volatile memory may be Random Access Memory (RAM), which acts as an external cache. By way of illustration and not limitation, many forms of RAM are available, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synchlink DRAM, SLDRAM) And direct memory bus random access memory (Direct Rambus RAM, DRRAM). The first memory 902 of the systems and methods described herein is intended to include, but not be limited to, these and any other suitable types of memory.

The first processor 903 may be an integrated circuit chip with signal processing capability. In the implementation process, each step of the above-mentioned method may be completed by an integrated logic circuit of hardware in the first processor 903 or an instruction in the form of software. The above-mentioned first processor 903 can be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a ready-made programmable gate array (Field Programmable Gate Array, FPGA) Or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of this application can be implemented or executed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in conjunction with the embodiments of the present application may be directly embodied as executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers and other storage media mature in the art. The storage medium is located in the first memory 902, and the first processor 903 reads the information in the first memory 902, and completes the steps of the above method in combination with its hardware.

It will be appreciated that the embodiments described herein may be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more Application Specific Integrated Circuits (ASIC), Digital Signal Processing (DSP), Digital Signal Processing Device (DSP Device, DSPD), programmable Logic Device (Programmable Logic Device, PLD), Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA), General Purpose Processor, Controller, Microcontroller, Microprocessor, Others for performing the functions described in this application electronic unit or a combination thereof. For a software implementation, the techniques described herein may be implemented through modules (eg, procedures, functions, etc.) that perform the functions described herein. Software codes may be stored in memory and executed by a processor. The memory can be implemented in the processor or external to the processor.

Optionally, as another embodiment, the first processor 903 is further configured to execute the method described in any one of the foregoing embodiments when running the computer program.

This embodiment provides an encoder, and the encoder may include a first determination unit, a first judgment unit, and a first prediction unit. In this way, when the attribute prediction is performed according to the geometric spatial relationship between the current node and the neighbor nodes, due to the expansion of the neighbor search range, the spatial correlation of the point cloud can be fully utilized, so that the prediction residual is smaller, and the code rate can be reduced. Thus, the coding efficiency is improved.

In yet another embodiment of the present application, based on the same inventive concept as the foregoing embodiments, see FIG. 10 , which shows a schematic structural diagram of a decoder 100 provided by an embodiment of the present application. As shown in FIG. 10, the decoder 100 may include: a second determination unit 1001, a second determination unit 1002, and a second prediction unit 1003; wherein,

The second determining unit 1001 is configured to determine the N neighbor nodes to be searched for the current node;

The second judging unit 1002 is configured to obtain a first attribute reconstruction value of the coplanar neighbor node if there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched;

The second judging unit 1002 is further configured to obtain a second attribute reconstruction value of the collinear neighbor node if there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched ;

The second prediction unit 1003 is configured to determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.

In some embodiments, referring to FIG. 10 , the decoder 100 may further include a second dividing unit 1004 configured to perform spatial division on the point cloud to obtain at least one point;

The second determining unit 1001 is further configured to, based on the geometric position of the current node, determine the neighbor information of the current node from the at least one point; and determine the N neighbor information from the neighbor information of the current node Neighbor node to be searched.

In some embodiments, the second determining unit 1001 is specifically configured to determine three neighbor nodes that must have been decoded among the six neighbor nodes coplanar with the current node as the three neighbor nodes to be searched, and will be collinear with the current node. The three neighbor nodes that must have been decoded among the twelve neighbor nodes of the The n neighbor nodes that may have been decoded in are determined as n neighbor nodes to be searched, and n is an integer greater than 0 and less than 9; and the obtained (6+n) neighbor nodes to be searched are determined as the N neighbor nodes to be searched neighbor node.

In some embodiments, N is an integer greater than 6.

In some embodiments, referring to FIG. 10 , the decoder 100 may further include a second computing unit 1005;

The second determining unit 1001 is further configured to determine the reference point of the current node;

The second calculation unit 1005 is configured to perform difference calculation on the coordinate information between the reference point and the N neighbor nodes to be searched to obtain N coordinate difference values; and a method for calculating the N coordinate difference values frame code, and the Morton code of the N coordinate differences is stored in a preset look-up table;

The second determining unit 1001 is further configured to determine Morton codes of the N neighbor nodes to be searched according to the preset lookup table.

In some embodiments, the second calculation unit 1005 is further configured to calculate the Morton code of the current node;

The second determining unit 1001 is further configured to, according to the Morton code of the current node, determine the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node; and determine the neighbor node corresponding to the smallest Morton code Determined as the reference point.

In some embodiments, the second determining unit 1001 is specifically configured to determine the coordinate information of the current node according to the Morton code of the current node; and determine the neighbor information of the current node based on the coordinate information of the current node and selecting the smallest Morton code from the Morton codes of the at least one neighbor node to obtain the neighbor node corresponding to the smallest Morton code.

In some embodiments, the second determining unit 1001 is further configured to determine the Morton code of the reference point;

The second calculation unit 1005 is further configured to obtain the Morton codes of the N coordinate difference values from the preset lookup table; and to obtain the Morton code of the reference point and the N coordinate difference values The Morton code is calculated to obtain the Morton codes of the N neighbor nodes to be searched.

In some embodiments, the second determination unit 1002 is further configured to traverse a preset point cloud sequence, and determine whether the preset point cloud sequence has a Morton code of the first neighbor node to be searched;

The second determining unit 1001 is further configured to determine that the first neighbor node to be searched exists in the N neighbor nodes to be searched if the determination result is yes; and to determine the N neighbor nodes to be searched if the determination result is no The first neighbor node to be searched does not exist in the neighbor nodes; wherein, the first neighbor node to be searched is any one of the N neighbor nodes to be searched.

In some embodiments, referring to FIG. 10 , the decoder 100 may further include a second sorting unit 1006;

The second determining unit 1001 is further configured to determine the Morton code of at least one point in the point cloud including the current node;

The second sorting unit 1006 is configured to arrange the Morton codes in ascending order to obtain a Morton code sequence; and truncate a Morton code smaller than the current node's Morton code from the Morton code sequence Morton code, and obtain the preset point cloud sequence according to the intercepted Morton code.

In some embodiments, the second judgment unit 1002 is further configured to judge whether the Morton code of the first neighbor node to be searched is smaller than the Morton code of the current node;

In some embodiments, the second determining unit 1001 is further configured to, when there is a first neighbor node to be searched in the N neighbor nodes to be searched, if the first neighbor node to be searched is a coplanar neighbor of the current node node, then determine that there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched; if the first neighbor node to be searched is the collinear neighbor node of the current node, then determine Among the N neighbor nodes to be searched, there is a collinear neighbor node that is collinear with the current node and has been decoded.

In some embodiments, referring to FIG. 10 , the decoder 100 may further include a parsing unit 1007 configured to parse the code stream to obtain the value of N.

In some embodiments, referring to FIG. 10 , the decoder 100 may further include a second setting unit 1008 configured to set the number of reference neighbor nodes to K, where K is an integer greater than 0 and less than or equal to N.

In some embodiments, the parsing unit 1007 is further configured to parse the code stream to obtain the value of K, where K represents the number of reference neighbor nodes.

In some embodiments, the second judgment unit 1002 is further configured to judge whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched;

The second determining unit 1001 is further configured to obtain the first number of the K coplanar neighbor nodes if there are K coplanar neighbor nodes that are coplanar with the current node and have been decoded in the N neighbor nodes to be searched an attribute reconstruction value;

The second predicting unit 1003 is specifically configured to perform mean calculation on the first attribute reconstruction values of the K coplanar neighbor nodes, and determine the attribute predicted value of the current node.

In some embodiments, the second judging unit 1002 is further configured to judge whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched; There is no coplanar neighbor node that is coplanar with the current node and has been decoded in the searched neighbor nodes, then judge whether there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched ;

The second determining unit 1001 is further configured to obtain the first number of the K collinear neighbor nodes if there are K collinear neighbor nodes that are collinear with the current node and have been decoded in the N neighbor nodes to be searched Two attribute reconstruction value;

The second prediction unit 1003 is specifically configured to perform mean calculation on the second attribute reconstruction values of the K collinear neighbor nodes, and determine the attribute prediction value of the current node.

The second determining unit 1001 is further configured to obtain the first number of the M coplanar neighbor nodes if there are M coplanar neighbor nodes that are coplanar with the current node and have been decoded among the N neighbor nodes to be searched. An attribute reconstruction value, M is an integer greater than 0 and less than K;

The second judging unit 1002 is further configured to judge whether there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched;

The second determining unit 1001 is further configured to obtain the first number of the L collinear neighbor nodes if there are L collinear neighbor nodes that are collinear with the current node and have been decoded among the N neighbor nodes to be searched. Two attribute reconstruction values; wherein, L is an integer greater than 0 and less than or equal to (K-M);

The second prediction unit 1003 is specifically configured to perform weighted calculation on the M first attribute reconstruction values and the L second attribute reconstruction values, and determine the attribute prediction value of the current node.

In some embodiments, the second prediction unit 1003 is further configured to: if there is no coplanar neighbor node coplanar with the current node and decoded and there is no coplanar neighbor node with the current node in the N neighbor nodes to be searched If the collinear neighbor node is collinear and has been decoded, the Morton code before the Morton code located at the current node is determined from the Morton code sequence, and the attribute of the point corresponding to the previous Morton code is determined. The reconstructed value is determined as the attribute prediction value of the current node.

It can be understood that, in this embodiment, a "unit" may be a part of a circuit, a part of a processor, a part of a program or software, etc., of course, it may also be a module, and it may also be non-modular. Moreover, each component in this embodiment may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of software function modules.

If the integrated unit is implemented in the form of a software functional module and is not sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such understanding, this embodiment provides a computer storage medium, which is applied to the decoder 100, where the computer storage medium stores a computer program, and when the computer program is executed by the second processor, any one of the foregoing embodiments is implemented the method described.

Based on the composition of the above-mentioned decoder 100 and the computer storage medium, see FIG. 11 , which shows a schematic diagram of a specific hardware structure of the decoder 100 provided by the embodiment of the present application. As shown in FIG. 11 , it may include: a second communication interface 1101 , a second memory 1102 and a second processor 1103 ; the various components are coupled together through a second bus system 1104 . It can be understood that the second bus system 1104 is used to realize the connection communication between these components. In addition to the data bus, the second bus system 1104 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, the various buses are labeled as the second bus system 1104 in FIG. 11 . in,

The second communication interface 1101 is used for receiving and sending signals in the process of sending and receiving information with other external network elements;

a second memory 1102 for storing computer programs that can be executed on the second processor 1103;

The second processor 1103 is configured to, when running the computer program, execute:

Determine the N neighbor nodes to be searched for the current node;

Optionally, as another embodiment, the second processor 1103 is further configured to execute the method described in any one of the foregoing embodiments when running the computer program.

It can be understood that the hardware functions of the second memory 1102 and the first memory 902 are similar, and the hardware functions of the second processor 1103 and the first processor 903 are similar; details are not described here.

This embodiment provides a decoder, and the decoder may include a second determination unit, a second judgment unit, and a second prediction unit. In this way, when the attribute prediction is performed according to the geometric spatial relationship between the current node and the neighbor nodes, due to the expansion of the neighbor search range, the spatial correlation of the point cloud can be fully utilized, so that the prediction residual is smaller, and the code rate can be reduced. Thereby, the decoding efficiency is improved.

It should be noted that, in this application, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements , but also other elements not expressly listed or inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages or disadvantages of the embodiments.

The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined under the condition of no conflict to obtain new method embodiments.

The features disclosed in the several product embodiments provided in this application can be combined arbitrarily without conflict to obtain a new product embodiment.

The features disclosed in several method or device embodiments provided in this application can be combined arbitrarily without conflict to obtain new method embodiments or device embodiments.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited to this. should be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Industrial Applicability

In the embodiment of the present application, on the encoder side, the N neighbor nodes to be searched of the current node are determined; if there is a coplanar neighbor node that is coplanar with the current node and has been coded among the N neighbor nodes to be searched, obtain the The first attribute reconstruction value of the coplanar neighbor node; if there is a collinear neighbor node that is collinear with the current node and has been encoded in the N neighbor nodes to be searched, then the second attribute reconstruction value of the collinear neighbor node is obtained; based on The first attribute reconstruction value and/or the second attribute reconstruction value determine the attribute prediction value of the current node. On the decoder side, by determining the N neighbor nodes to be searched of the current node; if there is a coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, obtain the coplanar neighbor node. The first attribute reconstruction value of the neighbor node; if there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the collinear neighbor node ; determining an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value. In this way, when the attribute prediction is performed according to the geometric spatial relationship between the current node and the neighbor nodes, due to the expansion of the neighbor search range, the spatial correlation of the point cloud can be fully utilized, so that the prediction residual is smaller, and the code rate can be reduced. Thus, the encoding and decoding efficiency is improved.

Claims

A prediction method, applied to an encoder, the method comprising:

Determine the N neighbor nodes to be searched for the current node;

If there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched, acquiring the first attribute reconstruction value of the coplanar neighbor node;

If there is a collinear neighbor node that is collinear with the current node and has been encoded in the N neighbor nodes to be searched, acquiring the second attribute reconstruction value of the collinear neighbor node;

An attribute prediction value of the current node is determined based on the first attribute reconstruction value and/or the second attribute reconstruction value.
The method according to claim 1, wherein the determining the N neighbor nodes to be searched for the current node comprises:

Perform spatial division on the point cloud to obtain at least one point;

determining the neighbor information of the current node from the at least one point based on the geometric position of the current node;

From the neighbor information of the current node, the N neighbor nodes to be searched are determined.
The method according to claim 2, wherein the neighbor information of the current node comprises: six neighbor nodes coplanar with the current node, twelve neighbor nodes colinear with the current node, and The eight neighbor nodes that the current node has in common.
The method according to claim 2, wherein, from the neighbor information of the current node, determining the N neighbor nodes to be searched, comprising:

The three neighbor nodes that must have been encoded in the six neighbor nodes that are coplanar with the current node are determined as three neighbor nodes to be searched, and the twelve neighbor nodes that are collinear with the current node must be encoded. Three neighbor nodes are determined as three neighbor nodes to be searched;

Based on the geometric space relationship between the current node and the neighbor nodes, n neighbor nodes that may have been encoded in the remaining nine neighbor nodes collinear with the current node are determined as the n neighbor nodes to be searched, where n is greater than an integer of 0 and less than 9;

The obtained (6+n) neighbor nodes to be searched are determined as the N neighbor nodes to be searched.
The method of claim 1, wherein N is an integer greater than 6.
The method according to claim 1, wherein after said determining the N neighbor nodes to be searched for the current node, the method further comprises:

determining the reference point of the current node;

Perform difference calculation on the coordinate information between the reference point and the N neighbor nodes to be searched to obtain N coordinate difference values;

calculating the Morton codes of the N coordinate differences, and storing the Morton codes of the N coordinate differences in a preset look-up table;

Morton codes of the N neighbor nodes to be searched are determined according to the preset lookup table.
The method according to claim 6, wherein the determining the reference point of the current node comprises:

calculating the Morton code of the current node;

According to the Morton code of the current node, determine the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node;

The neighbor node corresponding to the minimum Morton code is determined as the reference point.
The method according to claim 7, wherein the determining the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node according to the Morton code of the current node comprises:

Determine the coordinate information of the current node according to the Morton code of the current node;

determining the Morton code of at least one neighbor node in the neighbor information of the current node based on the coordinate information of the current node;

A minimum Morton code is selected from Morton codes of the at least one neighbor node to obtain a neighbor node corresponding to the minimum Morton code.
The method according to claim 6, wherein the determining the Morton codes of the N neighbor nodes to be searched according to the preset lookup table comprises:

determining a Morton code for the reference point;

From the preset lookup table, obtain the Morton codes of the N coordinate differences;

The Morton code of the reference point and the Morton code of the N coordinate differences are calculated to obtain the Morton codes of the N neighbor nodes to be searched.
The method according to claim 6, wherein after the determining the Morton codes of the N neighbor nodes to be searched, the method further comprises:

Traverse the preset point cloud sequence, and determine whether the preset point cloud sequence has a Morton code of the first neighbor node to be searched;

If the judgment result is yes, it is determined that the first neighbor node to be searched exists in the N neighbor nodes to be searched;

If the judgment result is no, it is determined that the first neighbor node to be searched does not exist in the N neighbor nodes to be searched;

Wherein, the first neighbor node to be searched is any neighbor node to be searched among the N neighbor nodes to be searched.
The method of claim 10, wherein the method further comprises:

determining a Morton code of at least one point in the point cloud comprising the current node;

Arrange the Morton codes in ascending order to obtain a Morton code sequence;

A Morton code smaller than the Morton code of the current node is intercepted from the Morton code sequence, and the preset point cloud sequence is obtained according to the intercepted Morton code.
The method according to claim 6, wherein after the determining the Morton codes of the N neighbor nodes to be searched, the method further comprises:

Determine whether the Morton code of the first neighbor node to be searched is smaller than the Morton code of the current node;

If the judgment result is yes, it is determined that the first neighbor node to be searched exists in the N neighbor nodes to be searched;

If the judgment result is no, it is determined that the first neighbor node to be searched does not exist in the N neighbor nodes to be searched;

Wherein, the first neighbor node to be searched is any neighbor node to be searched among the N neighbor nodes to be searched.
The method according to claim 10 or 12, wherein, when the first neighbor node to be searched exists in the N neighbor nodes to be searched, the method further comprises:

If the first neighbor node to be searched is a coplanar neighbor node of the current node, it is determined that there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched;

If the first neighbor node to be searched is a collinear neighbor node of the current node, it is determined that there is a collinear neighbor node that is collinear with the current node and has been coded among the N neighbor nodes to be searched.
The method of claim 1, wherein the method further comprises:

Write the value of N into the code stream.
The method according to any one of claims 1 to 14, wherein the method further comprises:

Set the number of reference neighbor nodes to K, where K is an integer greater than 0 and less than or equal to N.
The method of claim 15, wherein the method further comprises:

Judging whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched;

If there are K coplanar neighbor nodes that are coplanar with the current node and have been encoded in the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the K coplanar neighbor nodes;

Correspondingly, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value includes:

Perform mean calculation on the first attribute reconstruction values of the K coplanar neighbor nodes to determine the attribute prediction value of the current node.
The method of claim 15, wherein the method further comprises:

Judging whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched;

If there is no coplanar neighbor node that is coplanar with the current node and has been coded in the N neighbor nodes to be searched, it is determined whether there is a coplanar neighbor node that is collinear with the current node and has been coded in the N neighbor nodes to be searched. encoded collinear neighbor nodes;

If there are K collinear neighbor nodes that are collinear with the current node and have been coded in the N neighbor nodes to be searched, acquiring the second attribute reconstruction value of the K collinear neighbor nodes;

Correspondingly, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value includes:

Perform mean calculation on the second attribute reconstruction values of the K collinear neighbor nodes to determine the attribute prediction value of the current node.
The method of claim 15, wherein the method further comprises:

Judging whether there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched;

If there are M coplanar neighbor nodes that are coplanar with the current node and have been encoded among the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the M coplanar neighbor nodes, where M is greater than 0 and an integer less than K;

Judging whether there is a collinear neighbor node that is collinear with the current node and has been coded in the N neighbor nodes to be searched;

If there are L collinear neighbor nodes that are collinear with the current node and have been coded in the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the L collinear neighbor nodes; wherein, L is an integer greater than 0 and less than or equal to (K-M);

Correspondingly, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value includes:

A weighted calculation is performed on the M first attribute reconstruction values and the L second attribute reconstruction values to determine the attribute prediction value of the current node.
The method of claim 15, wherein the method further comprises:

Write the value of K into the code stream.
The method of claim 11, wherein the method further comprises:

If there is no coplanar neighbor node that is coplanar with the current node and has been coded and there is no colinear neighbor node that is colinear with the current node and has been coded in the N neighbor nodes to be searched, then In the Morton code sequence, the previous Morton code of the Morton code located at the current node is determined, and the attribute reconstruction value of the point corresponding to the previous Morton code is determined as the attribute prediction value of the current node.
A prediction method, applied to a decoder, the method comprising:

Determine the N neighbor nodes to be searched for the current node;

If there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched, acquiring the reconstruction value of the first attribute of the coplanar neighbor node;

If there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched, acquiring the second attribute reconstruction value of the collinear neighbor node;

An attribute prediction value of the current node is determined based on the first attribute reconstruction value and/or the second attribute reconstruction value.
The method according to claim 21, wherein the determining the N neighbor nodes to be searched for the current node comprises:

Perform spatial division on the point cloud to obtain at least one point;

determining the neighbor information of the current node from the at least one point based on the geometric position of the current node;

From the neighbor information of the current node, the N neighbor nodes to be searched are determined.
The method according to claim 22, wherein the neighbor information of the current node comprises: six neighbor nodes coplanar with the current node, twelve neighbor nodes colinear with the current node, and The eight neighbor nodes that the current node has in common.
The method according to claim 22, wherein, from the neighbor information of the current node, determining the N neighbor nodes to be searched, comprising:

Three neighbor nodes that must have been decoded in the six neighbor nodes that are coplanar with the current node are determined as three neighbor nodes to be searched, and the twelve neighbor nodes that are collinear with the current node must have been decoded. Three neighbor nodes are determined as three neighbor nodes to be searched;

Based on the geometric space relationship between the current node and neighbor nodes, n neighbor nodes that may have been decoded among the remaining nine neighbor nodes collinear with the current node are determined as n neighbor nodes to be searched, where n is greater than an integer of 0 and less than 9;

The obtained (6+n) neighbor nodes to be searched are determined as the N neighbor nodes to be searched.
The method of claim 21, wherein N is an integer greater than 6.
The method according to claim 21, wherein, after the determining the N neighbor nodes to be searched for the current node, the method further comprises:

determining the reference point of the current node;

Perform difference calculation on the coordinate information between the reference point and the N neighbor nodes to be searched to obtain N coordinate difference values;

calculating the Morton codes of the N coordinate differences, and storing the Morton codes of the N coordinate differences in a preset look-up table;

Morton codes of the N neighbor nodes to be searched are determined according to the preset lookup table.
The method according to claim 26, wherein the determining the reference point of the current node comprises:

calculating the Morton code of the current node;

According to the Morton code of the current node, determine the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node;

The neighbor node corresponding to the minimum Morton code is determined as the reference point.
The method according to claim 27, wherein the determining the neighbor node corresponding to the smallest Morton code from the neighbor information of the current node according to the Morton code of the current node comprises:

Determine the coordinate information of the current node according to the Morton code of the current node;

determining the Morton code of at least one neighbor node in the neighbor information of the current node based on the coordinate information of the current node;

A minimum Morton code is selected from Morton codes of the at least one neighbor node to obtain a neighbor node corresponding to the minimum Morton code.
The method according to claim 26, wherein the determining the Morton codes of the N neighbor nodes to be searched according to the preset lookup table comprises:

determining a Morton code for the reference point;

From the preset lookup table, obtain the Morton codes of the N coordinate differences;

The Morton code of the reference point and the Morton code of the N coordinate differences are calculated to obtain the Morton codes of the N neighbor nodes to be searched.
The method according to claim 26, wherein after the determining the Morton codes of the N neighbor nodes to be searched, the method further comprises:

Traverse the preset point cloud sequence, and determine whether the preset point cloud sequence has a Morton code of the first neighbor node to be searched;

If the judgment result is yes, it is determined that the first neighbor node to be searched exists in the N neighbor nodes to be searched;

If the judgment result is no, it is determined that the first neighbor node to be searched does not exist in the N neighbor nodes to be searched;

Wherein, the first neighbor node to be searched is any neighbor node to be searched among the N neighbor nodes to be searched.
The method of claim 30, wherein the method further comprises:

determining a Morton code of at least one point in the point cloud comprising the current node;

Arrange the Morton codes in ascending order to obtain a Morton code sequence;

A Morton code smaller than the Morton code of the current node is intercepted from the Morton code sequence, and the preset point cloud sequence is obtained according to the intercepted Morton code.
The method according to claim 26, wherein after the determining the Morton codes of the N neighbor nodes to be searched, the method further comprises:

Determine whether the Morton code of the first neighbor node to be searched is smaller than the Morton code of the current node;

If the judgment result is yes, it is determined that the first neighbor node to be searched exists in the N neighbor nodes to be searched;

If the judgment result is no, it is determined that the first neighbor node to be searched does not exist in the N neighbor nodes to be searched;

Wherein, the first neighbor node to be searched is any neighbor node to be searched among the N neighbor nodes to be searched.
The method according to claim 30 or 32, wherein, when the first neighbor node to be searched exists in the N neighbor nodes to be searched, the method further comprises:

If the first neighbor node to be searched is a coplanar neighbor node of the current node, it is determined that there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched;

If the first neighbor node to be searched is a collinear neighbor node of the current node, it is determined that there is a collinear neighbor node that is collinear with the current node and has been decoded among the N neighbor nodes to be searched.
The method of claim 21, wherein the method further comprises:

Parse the code stream and get the value of N.
The method of any one of claims 21 to 34, wherein the method further comprises:

Set the number of reference neighbor nodes to K, where K is an integer greater than 0 and less than or equal to N.
The method of any one of claims 21 to 34, wherein the method further comprises:

Parse the code stream to obtain the value of K; where K represents the number of reference neighbor nodes.
The method of claim 35 or 36, wherein the method further comprises:

Determine whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched;

If there are K coplanar neighbor nodes that are coplanar with the current node and have been decoded in the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the K coplanar neighbor nodes;

Correspondingly, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value includes:

Perform mean calculation on the first attribute reconstruction values of the K coplanar neighbor nodes to determine the attribute prediction value of the current node.
The method of claim 35 or 36, wherein the method further comprises:

Determine whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched;

If there is no coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, it is judged whether there is a coplanar neighbor node that is collinear with the current node and has been decoded among the N neighbor nodes to be searched. Decoded collinear neighbor nodes;

If there are K collinear neighbor nodes that are collinear with the current node and have been decoded in the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the K collinear neighbor nodes;

Correspondingly, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value includes:

Perform mean calculation on the second attribute reconstruction values of the K collinear neighbor nodes to determine the attribute prediction value of the current node.
The method of claim 35 or 36, wherein the method further comprises:

Determine whether there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched;

If there are M coplanar neighbor nodes that are coplanar with the current node and have been decoded among the N neighbor nodes to be searched, obtain the first attribute reconstruction value of the M coplanar neighbor nodes, where M is greater than 0 and an integer less than K;

Judging whether there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched;

If there are L collinear neighbor nodes that are collinear with the current node and have been decoded among the N neighbor nodes to be searched, obtain the second attribute reconstruction value of the L collinear neighbor nodes; wherein, L is an integer greater than 0 and less than or equal to (K-M);

Correspondingly, the determining the attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value includes:

A weighted calculation is performed on the M first attribute reconstruction values and the L second attribute reconstruction values to determine the attribute prediction value of the current node.
The method of claim 31, wherein the method further comprises:

If there is no coplanar neighbor node that is coplanar with the current node and has been decoded among the N neighbor nodes to be searched, and there is no colinear neighbor node that is collinear with the current node and has been decoded, then In the Morton code sequence, the previous Morton code of the Morton code located at the current node is determined, and the attribute reconstruction value of the point corresponding to the previous Morton code is determined as the attribute prediction value of the current node.
An encoder comprising a first determination unit, a first judgment unit and a first prediction unit; wherein,

The first determining unit is configured to determine the N neighbor nodes to be searched for the current node;

The first judgment unit is configured to obtain a first attribute reconstruction value of the coplanar neighbor node if there is a coplanar neighbor node that is coplanar with the current node and has been encoded in the N neighbor nodes to be searched ;

The first judging unit is further configured to obtain the second attribute reconstruction of the collinear neighbor node if there is a collinear neighbor node that is collinear with the current node and has been encoded in the N neighbor nodes to be searched value;

The first prediction unit is configured to determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.
An encoder comprising a first memory and a first processor; wherein,

the first memory for storing a computer program executable on the first processor;

The first processor is configured to execute the method according to any one of claims 1 to 20 when running the computer program.
A decoder comprising a second determination unit, a second judgment unit and a second prediction unit; wherein,

The second determining unit is configured to determine the N neighbor nodes to be searched for the current node;

The second judging unit is configured to obtain a first attribute reconstruction value of the coplanar neighbor node if there is a coplanar neighbor node that is coplanar with the current node and has been decoded in the N neighbor nodes to be searched ;

The second judgment unit is further configured to obtain the second attribute reconstruction of the collinear neighbor node if there is a collinear neighbor node that is collinear with the current node and has been decoded in the N neighbor nodes to be searched value;

The second prediction unit is configured to determine an attribute prediction value of the current node based on the first attribute reconstruction value and/or the second attribute reconstruction value.
A decoder comprising a second memory and a second processor; wherein,

the second memory for storing a computer program executable on the second processor;

The second processor is configured to execute the method according to any one of claims 21 to 40 when running the computer program.
A computer storage medium, wherein the computer storage medium stores a computer program, which when executed by a first processor implements the method according to any one of claims 1 to 20, or is executed by a second processor A method as claimed in any one of claims 21 to 40 is implemented when executed.