WO2021046817A1 - Point cloud data encoding and decoding method and system and storage medium - Google Patents

Abstract

A point cloud data encoding and decoding method and system and a storage medium. The encoding method comprises: performing hybrid encoding of position coordinates and attributes on input point cloud data (S110); and performing arithmetic encoding on a result of hybrid encoding to obtain a code stream of the point cloud data (S120), wherein the hybrid encoding comprises: performing multi-tree division on a space where the point cloud data is located to obtain multiple leaf nodes, and performing position encoding and attribute encoding on the point cloud data in the leaf nodes, and wherein the attribute data of one leaf node is located between the position data of the one leaf node and the position data of another leaf node in the code stream. The point cloud data encoding and decoding method and system and a storage medium use a hybrid encoding scheme of position coordinate encoding and attribute encoding, so that when it is necessary to use some of the point cloud data, it is not necessary to decode the position coordinates of all point cloud data and then decode attribute values, thereby improving the decoding efficiency.

Description

Point cloud data coding and decoding method, system and storage medium

Manual

Technical field

The present invention generally relates to the technical field of point cloud coding, and more specifically relates to a coding and decoding method, system and storage medium for point cloud data.

Background technique

A point cloud is a form of expression of a three-dimensional object or scene. It is composed of a set of discrete points that are randomly distributed in space and express the spatial structure and surface properties of the three-dimensional object or scene. In order to accurately reflect the information in the space, the number of discrete points required is huge. In order to reduce the bandwidth occupied by point cloud data storage and transmission, the point cloud data needs to be encoded and compressed.

In the point cloud data encoding and decoding process, the encoding and decoding of the position coordinates of the point cloud data points and the encoding and decoding of the attribute values are included, and the encoding and decoding of the position coordinates and the encoding and decoding of the attribute values are performed separately. Specifically, the existing point cloud data encoding method is: first encoding the position coordinates, and then encoding the attribute values after the encoding of the position coordinates is finished; the existing point cloud data decoding method is: first decoding the code stream of the position coordinates, After decoding all the position coordinate code streams, decode the attribute value code streams.

However, this encoding and decoding method has a problem: when the amount of data of the point cloud data point is large, there is a certain delay in the decoding process under this encoding and decoding scheme, and the decoder needs to decode all the position coordinates before it can be used. Decode the attribute value. When part of the point cloud data points need to be used, it is also necessary to decode the attribute values corresponding to each position coordinate one by one after decoding all the position coordinates, and then a complete and usable point cloud data point can be obtained. This obviously increases a lot of unnecessary calculations and makes efficiency low.

Summary of the invention

In order to solve the above-mentioned problems, the present invention has been proposed. The present invention provides a point cloud data coding and decoding scheme, which combines the characteristics of position coordinate coding and attribute coding, and adopts a coding scheme that combines position coordinate coding and attribute coding, so that when part of the point cloud data needs to be used, there is no need to After decoding the position coordinates of all the point cloud data, the attribute value decoding is performed, thereby improving the decoding efficiency. The following briefly describes the point cloud data encoding and decoding scheme proposed by the present invention, and more details will be described later in the specific implementation with reference to the accompanying drawings.

According to one aspect of the present invention, there is provided a point cloud data encoding method, the encoding method comprising: performing a hybrid encoding of position coordinates and attributes on the input point cloud data; and performing arithmetic encoding on the result of the hybrid encoding to obtain The code stream of the point cloud data; wherein, the hybrid encoding includes: performing multi-tree division on the space where the point cloud data is located to obtain multiple leaf nodes, and performing position encoding on the point cloud data in the leaf nodes, and Attribute coding; wherein the attribute data of a leaf node is located between the position data of the one leaf node and the position data of another leaf node in the code stream.

In an embodiment of the present invention, the encoding method further includes: quantizing the position coordinates of the input point cloud data before performing hybrid encoding of the position coordinates and attributes of the input point cloud data, and The hybrid coding is performed based on the quantized position coordinates.

In an embodiment of the present invention, the encoding method further includes: after quantizing the position coordinates of the input point cloud data, merging the attribute values of the point cloud data obtained by quantizing the same position coordinates; and Performing position coding and attribute coding on the point cloud data of any leaf node includes: coding the unique attribute value of the point cloud data in any leaf node.

In an embodiment of the present invention, the encoding method further includes: after quantizing the position coordinates of the input point cloud data, not merging the attribute values of the point cloud data obtained by quantizing the same position coordinates; In addition, performing position coding and attribute coding on the point cloud data of any leaf node includes: respectively coding the number and attribute values of the point cloud data in any leaf node.

In an embodiment of the present invention, the encoding method further includes: when any leaf node includes more than one point cloud data, combining the attribute values of the point cloud data in any leaf node; and The position encoding and attribute encoding of the point cloud data of any leaf node includes: encoding the merged attribute value of the point cloud data in the any leaf node.

In an embodiment of the present invention, the encoding method further includes: when any leaf node includes more than one point cloud data, not merging the attribute values of the point cloud data in any leaf node; and Performing position coding and attribute coding on the point cloud data of any leaf node includes: respectively coding the number and attribute values of the point cloud data in any leaf node.

In an embodiment of the present invention, the code stream includes an attribute merging flag bit, and the value of the attribute merging flag bit indicates whether the attribute values of the point cloud data are merged during the encoding process.

In an embodiment of the present invention, encoding the attribute value of the point cloud data includes: performing binary encoding on the attribute value of the point cloud data.

In an embodiment of the present invention, the performing binarization coding on the attribute value of the point cloud data includes: performing fixed-length coding, truncated Rice coding, or exponential Golomb coding on the attribute value of the point cloud data .

According to another aspect of the present invention, there is provided a method for decoding point cloud data. The decoding method includes: performing arithmetic decoding on a bitstream of the point cloud data to obtain an arithmetic decoding result; and performing position coordinates on the arithmetic decoding result and Hybrid decoding of attributes to obtain the respective position coordinates and attribute values of the point cloud data; wherein, the hybrid decoding includes: decoding the position coordinates based on multi-tree division to obtain multiple leaf nodes, wherein the attribute of one leaf node The data is located between the position data of the one leaf node and the position data of another leaf node in the code stream; position decoding and attribute decoding are performed on the point cloud data contained in the leaf node.

In an embodiment of the present invention, the decoding method further includes: determining, based on the code stream, whether the position coordinates of the point cloud data are quantized during the encoding process of the point cloud data, if so For quantization, inverse quantization is performed on the position coordinates obtained by the hybrid decoding.

In an embodiment of the present invention, the decoding method further includes: after quantizing the position coordinates of the point cloud data in the encoding process of the point cloud data, determining the encoding process of the point cloud data Whether the attribute values of the point cloud data with the same position coordinates obtained by quantization have been merged; if it is determined that the merge has been carried out, the position decoding and attribute decoding on the point cloud data of any leaf node includes: Decoding the unique attribute value of the point cloud data in the node; if it is determined that the merging is not performed, the position decoding and attribute decoding on the point cloud data of any leaf node includes: The number and attribute value are decoded separately.

In an embodiment of the present invention, the decoding method further includes: determining, based on the code stream, whether to correct the point cloud data in any leaf node that includes more than one point cloud data during the encoding process of the point cloud data. If it is determined to be merged, the position decoding and attribute decoding of the point cloud data of any leaf node includes: performing the merged attribute value of the point cloud data in any leaf node Decoding; if it is determined that no merging is performed, performing position decoding and attribute decoding on the point cloud data of any leaf node includes: decoding the number and attribute values of the point cloud data in any leaf node respectively.

In an embodiment of the present invention, the decoding method further includes: determining whether the attribute values of the point cloud data are merged during the encoding process based on the attribute merging flag in the code stream, and based on the determined As a result, position decoding and attribute decoding are performed on the leaf node.

In an embodiment of the present invention, decoding the attribute value of the point cloud data includes: performing binarization decoding on the attribute value of the point cloud data.

In an embodiment of the present invention, the performing binarization decoding on the attribute value of the point cloud data includes: performing fixed-length decoding, truncated Rice decoding, or exponential Columbus decoding on the attribute value of the point cloud data .

According to another aspect of the present invention, a point cloud data encoding system is provided, the encoding system includes a storage device and a processor, and the storage device stores a computer program run by the processor, and the computer program The point cloud data encoding method described in any one of the above is executed when being run by the processor.

According to another aspect of the present invention, a point cloud data decoding system is provided. The encoding system includes a storage device and a processor. The storage device stores a computer program run by the processor, and the computer program The point cloud data decoding method described in any one of the above is executed when being run by the processor.

According to another aspect of the present invention, a storage medium is provided, and a computer program is stored on the storage medium, and the computer program executes the point cloud data encoding method described in any one of the above items when the computer program is running.

According to another aspect of the present invention, there is provided a storage medium having a computer program stored on the storage medium, and the computer program executes the point cloud data decoding method described in any one of the above items when the computer program is running.

According to the point cloud data coding and decoding method, system and storage medium of the embodiment of the present invention, combining the characteristics of position coordinate coding and the characteristics of attribute coding, a coding scheme that combines position coordinate coding and attribute coding is adopted, so that part of the point cloud is used when needed. When data, it is not necessary to decode the position coordinates of all the point cloud data and then decode the attribute value, thereby improving the decoding efficiency.

Description of the drawings

Fig. 1 shows a schematic flowchart of a point cloud data encoding method according to an embodiment of the present invention;

Fig. 2 shows a schematic diagram of octree division according to an embodiment of the present invention;

Fig. 3 shows a schematic diagram of recursive division of an octree according to an embodiment of the present invention;

Fig. 4 shows a schematic block diagram of a point cloud data encoding system according to an embodiment of the present invention;

Fig. 5 shows a schematic flowchart of a point cloud data decoding method according to an embodiment of the present invention;

Fig. 6 shows a schematic block diagram of a point cloud data decoding system according to an embodiment of the present invention.

detailed description

In order to make the objectives, technical solutions, and advantages of the present invention more obvious, the exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited by the exemplary embodiments described herein. Based on the embodiments of the present invention described in the present invention, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present invention.

In the following description, a lot of specific details are given in order to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features known in the art are not described.

It should be understood that the present invention can be implemented in different forms and should not be construed as being limited to the embodiments presented here. On the contrary, the provision of these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the present invention to those skilled in the art.

The purpose of the terms used here is only to describe specific embodiments and not as a limitation of the present invention. When used herein, the singular forms "a", "an" and "the/the" are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "composition" and/or "including", when used in this specification, determine the existence of the described features, integers, steps, operations, elements and/or components, but do not exclude one or more other The existence or addition of features, integers, steps, operations, elements, components, and/or groups. As used herein, the term "and/or" includes any and all combinations of related listed items.

In order to thoroughly understand the present invention, detailed steps and detailed structures will be presented in the following description to explain the technical solutions proposed by the present invention. The preferred embodiments of the present invention are described in detail as follows. However, in addition to these detailed descriptions, the present invention may also have other embodiments.

As mentioned above, the current point cloud data encoding method encodes the attribute value of the point cloud data only after all the position coordinates of the point cloud data are encoded, so that all the position coordinates need to be decoded before the attribute value can be decoded. , When only part of the point cloud data points need to be used, the coding and decoding efficiency is low. Based on this, the present invention provides an encoding and decoding solution for point cloud data. The encoding and decoding solution for point cloud data according to embodiments of the present invention will be described below with reference to the accompanying drawings.

Fig. 1 shows a schematic flowchart of a point cloud data encoding method 100 according to an embodiment of the present invention. As shown in FIG. 1, the point cloud data encoding method 100 may include the following steps:

In step S110, perform hybrid encoding of position coordinates and attributes on the input point cloud data. The hybrid encoding includes: multi-tree division of the space where the point cloud data is located to obtain multiple leaf nodes, The point cloud data is position-encoded and attribute-encoded.

In the embodiment of the present invention, a distance measuring device such as a laser scanner and a lidar may be used to collect point cloud data for a certain object or a certain scene. Generally, the collected point cloud data includes position coordinates in a three-dimensional direction and attribute information of the position. For example, the position coordinates of each point cloud data can be expressed as (x, y, z), and the attribute information of each point cloud data can include attribute values such as color (R, G, B) or reflectance.

In the embodiment of the present invention, the point cloud data is subjected to mixed coding of position coordinates and attributes, that is, a mixed coding of position coordinate coding (also referred to as position coding for short) and attribute coding. Wherein, the encoding of the position coordinates may adopt multi-tree partition encoding (for example, octree partition encoding), and the encoding of the attributes is performed in the process of encoding the position coordinates. Specifically, multi-tree division is performed on the space where the point cloud data is located to obtain multiple leaf nodes (that is, the nodes of the last layer obtained by multi-tree division), and the leaf nodes may have a side length less than or equal to the set minimum side Long block. Position coding and attribute coding are performed on the point cloud data in the leaf node, and the attribute coding result of one leaf node is located between the position coding result of one leaf node and the position coding result of another leaf node. The following uses octree division coding as an example to detail the above-mentioned mixed coding of position coordinates and attributes on point cloud data proposed by the present invention.

The octree partition coding is a method of using octree partitioning and compressing the coordinate position. The division of each layer of the octree uses the coordinates of the center point of the current block to divide the current block into eight small sub-blocks through the center point. A schematic diagram of performing an octree division on a coding block is shown in Figure 2. The octree division can be implemented according to the following process: First, according to the maximum value of the position coordinates of the point cloud data (hereinafter referred to as the position coordinates) in each of the three directions, one of the three maximum values is obtained. The maximum value in the selected value is used to determine the side length of the initialization block used to initialize the octree division. Generally, the side length of the initialization block can be an integer power of 2 and be greater than or equal to and closest to the selected value. It should be understood that the selected value can be written into the header information of the code stream file for use by the decoding end. Based on the determination of the side length of the initialization block, the octree partition coding can be started. The division of each layer of the octree uses the coordinates of the center point of the current block for sub-block division, and the current block is divided into eight small sub-blocks through the center point. After the sub-block is divided, it will judge whether there is point cloud data in each sub-block, and the sub-blocks with point cloud data will be further divided until the sub-block is divided to the minimum, for example, the side length of the sub-block reaches the preset minimum side Long (usually set the preset minimum side length to 1), that is, stop dividing when the leaf node is obtained.

The schematic diagram of the recursive division of the octree is shown in Figure 3. When compressing and encoding the position coordinates of the point cloud data, according to the breadth-first traversal order of the octree, the encoding is performed layer by layer, and the division of each octree is encoded layer by layer. In addition, the eight sub-blocks obtained after the octree division of each block will determine whether it contains point cloud data points. If it contains point cloud data points, it will be further divided. Taking Figure 3 as an example, the black squares in the figure indicate that the current sub-block contains point cloud data points, and the white squares indicate that the current sub-block does not contain point cloud data points. When the root node is divided into an octree, it is determined in turn whether there are point cloud data points in each small block. Take the first layer in Figure 3 as an example, the third block in the first divided octree contains For point cloud data points, the remaining seven blocks do not contain point cloud data points, so this division can be represented by 8 bits, specifically 0010 0000. Then divide the block containing the data points, that is, the second level of division in Figure 3. As a result, the third block and the eighth block contain point cloud data points this time, then this octree division can be used 0010 0001. Then divide the two sub-blocks of cloud data points in turn. It is further divided in the order of breadth first. In the third layer, the left sub-block is first divided, and the result of the division is shown in the left half of the third layer in Figure 3. The first and fourth sub-blocks contain point cloud data points, then this octree division It can be represented by 1001 0000. Then divide the second sub-block of the second layer of cloud data points. The result of the division is shown in the right half of the third layer in Figure 3. The second and eighth sub-blocks contain point cloud data points. Then this octree division can be represented by 0100 0001. The division of the latter layers is similar to the previous description, and is divided layer by layer in the order of breadth first traversal. The binary code stream corresponding to the division result in Figure 3 is 0010 0000 0010 0001 1001 0000 0100 0001.... When it is divided to the last layer, that is, the side length of the sub-block reaches the preset minimum side length, the deepest position of the octree division has been reached at this time, that is, the leaf node block of the current octree has been reached, and no further division is required According to the embodiment of the present invention, at this time, the number of point cloud data points contained in the current leaf node block and the respective attribute values of the one or more point cloud data points can be encoded.

In the embodiment of the present invention, when the leaf node block contains a point cloud data point, it can be directly encoded with a 0 to represent it, and then the attribute value corresponding to this position coordinate will be encoded; when the leaf node block contains more than one point cloud For data points, suppose that there are n point cloud data points in the current leaf node block. At this time, a 1 will be coded, then the value (n-1) will be coded, and then the attribute values corresponding to the n point cloud data points will be coded continuously . The number of point cloud points contained in each leaf node block and the corresponding attribute value are sequentially coded to obtain the result of the mixed coding of the position coordinates and attributes of the point cloud data.

In the embodiment of the present invention, the encoding of the attribute value of the point cloud data may be binary encoding. Exemplarily, the encoding of the attribute value of the point cloud data may be fixed-length encoding, truncated Rice encoding, or exponential Golomb encoding. Of course, this is only exemplary, and any other suitable method may be used to encode the attribute value of the point cloud data. The following describes the above three attribute encoding methods in detail.

In an embodiment of the present invention, the attribute value of the point cloud data may be binarized by using a fixed-length code scheme. In an example, the bit depth of the fixed-length code scheme of the attribute value can be determined according to the largest value among the attribute values of all the point cloud data. In another example, the point cloud data acquisition device may also determine the bit depth of the fixed-length code scheme of the attribute value, that is, determine the bit depth of the fixed-length code scheme of the attribute value according to the bit depth of the acquired attribute value . When encoding the attribute value of any point cloud data of any leaf node block, the attribute value of the point cloud data can be converted into a binary number with a bit width equal to the above determined bit depth to realize the attribute value of the point cloud data The binarization.

In another embodiment of the present invention, the attribute value of the point cloud data may be binarized by using a truncated Rice scheme. In an example, the threshold value can be set to cMax, the Rice parameter is R, and the attribute value is Ref. The truncated Rice code is formed by concatenating the prefix code and the suffix code. The prefix code is generated according to the prefix value P, and the calculation method of the prefix value P is P=Ref>>R. Through calculation, if the prefix value P is less than the value (cMax＞＞R), the prefix code is composed of P 1s and one 0, and the length is P+1; if the prefix value P is greater than or equal to the value (cMax＞＞R), the prefix The code consists of (cMax>>R) 1s, and the length is (cMax>>R). The suffix code is generated according to the suffix value S: when the attribute value Ref is less than cMax, the calculation method of the suffix value S is S=Ref-(P＜＜R), the suffix code is a binary string of S, and the length is R; When the attribute value Ref is greater than or equal to cMax, there is no suffix code. According to this, the attribute value of any point cloud data of any leaf node block can be binarized.

In another embodiment of the present invention, the attribute value of the point cloud data may be binarized by adopting a k-order exponential Golomb code scheme. The k-order exponential Golomb code consists of a prefix and a suffix. Both the prefix and the suffix depend on the order k of the exponential Golomb code. The k-order exponential Golomb code used to represent the non-negative integer attribute value Ref can be generated by the following steps: (1) Write the attribute value Ref in binary form, remove the lowest k bits, and then add 1; (2) Calculate and leave For the number of bits under, subtract 1 from this number, that is, the number of prefix zeros that need to be added; (3) The lowest k bits removed in step (1) are added to the end of the bit string. The following describes the attribute value Ref = 4, the first-order exponential Golomb code (ie k = 1) as an example: (1) The binary representation of 4 is 100, and after removing the lowest 1 bit, 0 becomes 10, and after adding 1, it becomes 11; (2) The number of bits of 11 is 2, so the number of 0s in the prefix is 1; (3) The lowest bit of the bit string is filled with the 0 removed in step (1), and the final codeword is 0110. For the k-order exponential Golomb code, the prefix is composed of m consecutive 0s and one 1, and the suffix is composed of m+k, which is the binary representation ^{of Ref-2 k} (2 ^{m -1).} According to this, the attribute value of any point cloud data of any leaf node block can be binarized.

As mentioned above, the point cloud data in the leaf node is encoded with position and attribute, and the attribute encoding result of one leaf node is located between the position encoding result of one leaf node and the position encoding result of another leaf node, namely The result of hybrid coding of the position coordinates and attributes of the point cloud data is obtained.

The above exemplarily describes the process of step 110 of the point cloud data encoding method 100 according to the embodiment of the present invention. The subsequent steps of the encoding method 100 are described below with reference to FIG. 1.

In step 120, arithmetic coding is performed on the result of the hybrid coding to obtain a code stream of point cloud data, wherein the attribute data of one leaf node is located in the code stream at the position data of the one leaf node and the other leaf node Between the location data.

In the embodiment of the present invention, the mixed coding result (binary code stream) obtained in step S120 can be sent to the arithmetic coding engine for arithmetic coding, then the position coordinates and attribute code streams of the point cloud data can be obtained, and the point-matching can be realized. The position coordinate coding and attribute coding of cloud data.

In a further embodiment of the present invention, the encoding method 100 may further include (not shown in FIG. 1): before performing mixed encoding of the position coordinates and attributes of the input point cloud data, performing the mixed encoding of the input point The position coordinates of the cloud data are quantized, and the hybrid coding is performed based on the quantized position coordinates.

Exemplarily, the difference between the maximum value and the minimum value in each of the three directions (ie, the x-axis direction, the y-axis direction, and the z-axis direction) according to the position coordinates of the input point cloud data, and According to the input quantization precision parameter, the position coordinates of each input point cloud data are quantized to simplify the coding operation of the position coordinates of the point cloud data. Exemplarily, the position coordinates of the input point cloud data can be converted into integer coordinates greater than or equal to zero by quantization.

Further, after quantifying the position coordinates of all point cloud data, duplicate coordinate values may appear. At this time, the same position coordinate corresponds to multiple point cloud data, that is, the same position coordinate may correspond to multiple attributes value. Based on this, in an example, after the position coordinates of the input point cloud data are quantified, the attribute values of the point cloud data obtained by quantizing the same position coordinates may be merged. Among them, the combination of attribute values may be, for example, performing a weighted summation, averaging, or other suitable operations on multiple attribute values. In short, multiple attribute values are converted into one attribute value. In this example, since the attribute values of the point cloud data at the same position coordinates after quantization are merged, each leaf node obtained after the multi-tree division can be regarded as including only one point cloud data point. The attribute value is also one. Because although the leaf node may include multiple point cloud data points, its position coordinates are the same, so the position is regarded as one point (position encoding is based on including a point cloud data point for position encoding, that is, as mentioned above, the leaf The node includes a point cloud data point code 0), and its attribute values are merged, so the attribute code is to encode the unique attribute value. Further, after the above analysis, in this example, the leaf node always includes a point cloud data point and an attribute value, so the number of point cloud data points may not be encoded when encoding (because the number of point cloud data points in each leaf node The number of point cloud data is 1), and only the unique attribute value of the point cloud data point in each leaf node can be encoded, which can further reduce the number of bit stream bits and save encoding overhead. In addition, in order to make the decoder know that when decoding leaf nodes, only the attribute values need to be decoded without the need to decode the number of point cloud data points, the attribute merging flag can be set in the code stream, and this flag can be used to identify the point in the encoding process. The attribute values of cloud data have been merged. Of course, the attribute values of the point cloud data can also be merged by default in the encoding process without setting the attribute merge flag bit.

In another example, after the position coordinates of the input point cloud data are quantified, the attribute values of the point cloud data obtained by quantizing the same position coordinates may not be merged. In this example, since the attribute values of the point cloud data at the same position coordinates are not merged after quantization, the leaf nodes obtained after the multi-tree division may include the situation where there are multiple point cloud data points in the leaf nodes. And each point cloud data point has its own attribute value. In this case, the encoding of the point cloud data in the leaf node may include encoding the number of point cloud data in the leaf node and separately encoding the attribute value of each point cloud data. Similarly, in order to make the decoding end understand that the number of point cloud data points and attribute values need to be decoded when decoding leaf nodes, the attribute merging flag can be set in the code stream, and the flag can be used to identify the point cloud during the encoding process. The attribute values of the data are not merged. Of course, it is also possible to default the encoding process not to merge the attribute values of the point cloud data without setting the attribute merging flag bit.

In a further embodiment of the present invention, the encoding method 100 may further include (not shown in FIG. 1): when any leaf node includes more than one point cloud data, combining the point cloud data in any leaf node The attribute values of the data are merged; and the position encoding and attribute encoding of the point cloud data of any leaf node includes: encoding the merged attribute value of the point cloud data in any leaf node. In this embodiment, the attribute value of each point cloud data in any leaf node that includes more than one point cloud data is merged, so that by default, each leaf node includes one point cloud data and one attribute value, so the leaf node When encoding, it is sufficient to encode only the unique attribute value of the leaf node (for example, the merged attribute value, referred to as the merged attribute value), without the need to encode the number of point cloud data in the leaf node, thereby reducing the bit stream bit. Save coding overhead. In addition, in order to make the decoder know that when decoding leaf nodes, only the attribute values need to be decoded without the need to decode the number of point cloud data points, the attribute merging flag can be set in the code stream, and this flag can be used to identify the point in the encoding process. The attribute values of cloud data have been merged. Of course, the attribute values of the point cloud data can also be merged by default in the encoding process without setting the attribute merge flag bit.

In another embodiment of the present invention, the encoding method 100 may further include (not shown in FIG. 1): when any leaf node includes more than one point cloud data, the Merging the attribute values of the point cloud data; and performing position encoding and attribute encoding on the point cloud data of any leaf node includes: respectively encoding the number and attribute values of the point cloud data in any leaf node . In this embodiment, the attribute value of each point cloud data in any leaf node that includes more than one point cloud data is not merged, so when encoding the leaf node, it is still necessary to encode the number and points of the point cloud data in the leaf node. The attribute value of the cloud data. Similarly, in order to make the decoding end understand that the number of point cloud data points and attribute values need to be decoded when decoding leaf nodes, the attribute merging flag can be set in the code stream, and the flag can be used to identify the point cloud during the encoding process. The attribute values of the data are not merged. Of course, it is also possible to default the encoding process not to merge the attribute values of the point cloud data without setting the attribute merging flag bit.

The above exemplarily describes the point cloud data encoding method according to the embodiment of the present invention. Based on the above description, the point cloud data coding method according to the embodiment of the present invention combines the characteristics of position coordinate coding and the characteristics of attribute coding, and adopts a coding scheme that combines position coordinate coding and attribute coding, so that part of the point cloud data is used when needed. In this case, it is not necessary to decode the position coordinates of all the point cloud data and then decode the attribute values, thereby improving the decoding efficiency. Exemplarily, the method for encoding point cloud data according to an embodiment of the present invention may be implemented in a device, device, or system having a memory and a processor.

The following describes a point cloud data encoding system according to an embodiment of the present invention with reference to FIG. 4. Fig. 4 shows a schematic block diagram of a point cloud data encoding system 400 according to an embodiment of the present invention. The point cloud data encoding system 400 includes a storage device 410 and a processor 420.

Wherein, the storage device 410 stores a program for implementing the corresponding steps in the method for encoding point cloud data according to an embodiment of the present invention. The processor 420 is configured to run a program stored in the storage device 410 to execute corresponding steps of the point cloud data encoding method according to the embodiment of the present invention.

In one embodiment, when the program is executed by the processor 420, the point cloud data encoding system 400 is caused to perform the following steps: perform a hybrid encoding of position coordinates and attributes on the input point cloud data; The result is arithmetic coding to obtain the code stream of the point cloud data; wherein, the hybrid coding includes: multi-tree division of the space where the point cloud data is located to obtain a plurality of leaf nodes, and the point cloud in the leaf nodes The data is subjected to position coding and attribute coding; wherein, the attribute data of one leaf node is located between the position data of the one leaf node and the position data of another leaf node in the code stream.

In an embodiment of the present invention, when the program is run by the processor 420, the point cloud data encoding system 400 is also caused to perform the following steps: before the input point cloud data is mixed encoding of position coordinates and attributes Quantifying the position coordinates of the input point cloud data, and performing the hybrid coding based on the quantized position coordinates.

In an embodiment of the present invention, when the program is run by the processor 420, the point cloud data encoding system 400 is also caused to perform the following steps: after quantizing the position coordinates of the input point cloud data, The attribute values of the point cloud data with the same position coordinates are quantified and merged; and the position coding and attribute coding of the point cloud data of any leaf node includes: performing the unique attribute value of the point cloud data in any leaf node coding.

In an embodiment of the present invention, when the program is run by the processor 420, the point cloud data encoding system 400 is also caused to perform the following steps: after quantizing the position coordinates of the input point cloud data, Quantifying the attribute values of the point cloud data with the same position coordinates are not merged; and performing position coding and attribute coding on the point cloud data of any leaf node includes: summing the number of point cloud data in any leaf node The attribute values are coded separately.

In an embodiment of the present invention, when the program is run by the processor 420, the point cloud data encoding system 400 is also caused to perform the following steps: when any leaf node includes more than one point cloud data, the any Merging the attribute values of the point cloud data in a leaf node; and performing position encoding and attribute coding on the point cloud data of any leaf node includes: merging attribute values of the point cloud data in any leaf node Encode.

In an embodiment of the present invention, when the program is run by the processor 420, the point cloud data encoding system 400 is also caused to perform the following steps: when any leaf node includes more than one point cloud data, the Merging the attribute values of the point cloud data in any leaf node; and performing position encoding and attribute encoding on the point cloud data of any leaf node includes: the number of point cloud data in any leaf node And the attribute value are coded separately.

In an embodiment of the present invention, the code stream includes an attribute merging flag bit, and the value of the attribute merging flag bit indicates whether the attribute values of the point cloud data are merged during the encoding process.

In an embodiment of the present invention, when the program is executed by the processor 420, the point cloud data encoding system 400 is executed to encode the attribute value of the point cloud data, including: encoding the point cloud data The attribute value of is binary coded.

In an embodiment of the present invention, when the program is executed by the processor 420, the point cloud data encoding system 400 executes the binary encoding of the attribute values of the point cloud data, including: The attribute values of the point cloud data are subjected to fixed-length coding, truncated Rice coding or exponential Golomb coding.

In addition, according to an embodiment of the present invention, there is also provided a storage medium on which program instructions are stored, and the program instructions are used to execute the point cloud data in the embodiments of the present invention when the program instructions are run by a computer or a processor. The corresponding steps of the encoding method. The storage medium may include, for example, a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

In an embodiment, the computer program instructions can execute the point cloud data encoding method according to the embodiment of the present invention when being executed by a computer.

In one embodiment, the computer program instructions, when run by the computer or processor, cause the computer or processor to perform the following steps: perform hybrid encoding of position coordinates and attributes on the input point cloud data; and encode the hybrid encoding The result is arithmetic coding to obtain the code stream of the point cloud data; wherein, the hybrid coding includes: multi-tree division of the space where the point cloud data is located to obtain a plurality of leaf nodes, and the point cloud in the leaf nodes The data is subjected to position coding and attribute coding, wherein the attribute data of one leaf node is located between the position data of the one leaf node and the position data of another leaf node in the code stream.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor executes the following steps: before performing the mixed encoding of the position coordinates and attributes on the input point cloud data Quantifying the position coordinates of the input point cloud data, and performing the hybrid coding based on the quantized position coordinates.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor further executes the following steps: after quantifying the position coordinates of the input point cloud data, The attribute values of the point cloud data with the same position coordinates are quantified and merged; and the position coding and attribute coding of the point cloud data of any leaf node includes: performing the unique attribute value of the point cloud data in any leaf node coding.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor further executes the following steps: after quantifying the position coordinates of the input point cloud data, Quantifying the attribute values of the point cloud data with the same position coordinates are not merged; and performing position coding and attribute coding on the point cloud data of any leaf node includes: summing the number of point cloud data in any leaf node The attribute values are coded separately.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor also executes the following steps: when any leaf node includes more than one point cloud data, the Merging the attribute values of the point cloud data in a leaf node; and performing position encoding and attribute coding on the point cloud data of any leaf node includes: merging attribute values of the point cloud data in any leaf node Encode.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor also executes the following steps: when any leaf node includes more than one point cloud data, the Merging the attribute values of the point cloud data in any leaf node; and performing position encoding and attribute encoding on the point cloud data of any leaf node includes: the number of point cloud data in any leaf node And the attribute value are coded separately.

In an embodiment of the present invention, the code stream includes an attribute merging flag bit, and the value of the attribute merging flag bit indicates whether the attribute values of the point cloud data are merged during the encoding process.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor executes the encoding of the attribute value of the point cloud data, including: The attribute value is binary coded.

In an embodiment of the present invention, when the computer program instructions cause the computer or the processor to execute the binary encoding of the attribute value of the point cloud data when being executed by the computer or the processor, the method includes: The attribute value of the point cloud data is coded with fixed length, truncated Rice coding or exponential Golomb coding.

According to another aspect of the present invention, a method for decoding point cloud data is also provided. Since the point cloud data decoding method according to the embodiment of the present invention corresponds to the point cloud data encoding method according to the embodiment of the present invention, for the sake of brevity, some processes in the decoding process that are similar or identical to the encoding process are no longer Go into details.

FIG. 5 shows a schematic flowchart of a method 500 for decoding point cloud data according to an embodiment of the present invention. As shown in FIG. 5, the method 500 for decoding point cloud data may include the following steps:

In step S510, arithmetic decoding is performed on the bit stream of the point cloud data to obtain the arithmetic decoding result.

In the embodiment of the present invention, step S510 corresponds to step S120 of the point cloud data encoding method 100 according to the embodiment. In this step, the code stream of the point cloud data is subjected to the inverse process of arithmetic encoding, that is, arithmetic decoding , Get the arithmetic decoding result.

In step S520, perform hybrid decoding of the position coordinates and attributes of the arithmetic decoding result to obtain the respective position coordinates and attribute values of the point cloud data; wherein, the hybrid decoding includes: performing position coordinates based on multitree division A plurality of leaf nodes are obtained by decoding, wherein the attribute data of one leaf node is located between the position data of the one leaf node and the position data of another leaf node in the code stream; Point cloud data performs position decoding and attribute decoding.

In the embodiment of the present invention, step S520 corresponds to step S110 of the point cloud data encoding method 100 according to the embodiment. In this step, the arithmetic decoding result obtained in step S520 is decoded with position coordinates and attributes. . First, the header information of the code stream can be decoded to obtain the maximum value of the maximum value of the position coordinates of the point cloud data in each of the three directions, and then based on this value, the edge of the initialization block used for multi-tree division can be determined long. Based on the determination of the side length of the initialization block, multitree division decoding (for example, octree division decoding) can be started. The following uses octree division decoding as an example to detail the above-mentioned hybrid decoding of the position coordinates and attributes of the arithmetic decoding result proposed by the present invention.

Decode 8 bits in turn to determine the octree division of a block. Each bit indicates whether a sub-block contains point cloud data points. If it is 1, it means it contains point cloud data points. If it is 0, it means it does not contain point cloud data points. The octree division and decoding are continued on the block containing the point cloud data points. When the octree division and decoding process proceeds until the side length of the block reaches the preset minimum side length (the preset minimum side length is usually set to 1), it indicates that the division has ended. Next, decode the number of point cloud points contained in each leaf node. When decoded to a 0, it indicates that the current leaf node block contains a point cloud data point, and then decode the attribute value corresponding to the position coordinate of the point cloud data point. When decoded to a 1, it indicates that the current leaf node block contains more than one point cloud point, and then decodes to obtain a value (n-1), which means that the current leaf node block contains n point cloud data points. Then, the n attribute values are continuously decoded, which are the attribute values corresponding to the position coordinates of the point cloud points in the leaf node block. Among them, the decoding numerical scheme should be a scheme corresponding to the encoding scheme for the attribute value adopted in the encoding process.

As mentioned above, in the embodiment of the present invention, the encoding of the attribute value of the point cloud data may be binary encoding. Exemplarily, the encoding of the attribute value of the point cloud data may be fixed-length encoding, truncated Rice encoding, or exponential Golomb encoding. Therefore, correspondingly, in the embodiment of the present invention, the decoding of the attribute value of the point cloud data may be binarization decoding. Exemplarily, the decoding of the attribute value of the point cloud data may be fixed-length decoding, truncated Rice decoding, or exponential Golomb decoding. It should be understood that this is only exemplary, and the decoding of the attribute value of the point cloud data may also be any other suitable method, depending on the encoding method adopted for the attribute value. The above three attribute decoding methods are described in detail below.

In an embodiment of the present invention, the attribute value of the point cloud data is binarized using a fixed-length code scheme. In an example, the header information in the code stream can be decoded to obtain the bit depth used by the fixed-length code binarization scheme, and the bit width binary code stream represented by the bit depth can be decoded to obtain the corresponding attribute value.

In another embodiment of the present invention, the attribute value of the point cloud data is binarized using a truncated Rice scheme. In an example, the header information in the code stream can be decoded to obtain the threshold value cMax and the Rice parameter R, and then the prefix code and the suffix code can be decoded. First decode the prefix code and decode it bit by bit. If it is decoded to a bit of 0, or if the number of decoded bits is greater than or equal to (cMax>>R), stop decoding the prefix code. If the number of prefix code bits is greater than or equal to (cMax>>R), the attribute value is cMax. Otherwise, the prefix code can be set to P 1s and one 0, and then R bits are decoded. The calculation of the attribute value Ref can be expressed by the formula: Ref=S+(P=R), based on which the truncated Rice code can be realized decoding.

In yet another embodiment of the present invention, the attribute value of the point cloud data is binarized using a k-order exponential Golomb code scheme. In an example, when decoding the k-order exponential Golomb code, the first non-zero bit is searched from the current position of the bit stream, and the number of zero bits found is recorded as m. After the first non-zero bit, m+ The decimal value of the k-bit binary bit string is value, and the calculated decoded value Ref can be expressed by a formula: Ref=2 ^m+k -2 ^k +Value, and the attribute value corresponding to this position coordinate can be decoded.

The number of point cloud points contained in each leaf node block and the corresponding attribute value are sequentially decoded, that is, the result of a mixed decoding of the position coordinates and attributes of the arithmetic decoding result is obtained.

In a further embodiment of the present invention, the decoding method 500 may further include (not shown in FIG. 5): determining, based on the code stream, whether to position the point cloud data during the encoding process of the point cloud data. The coordinates are quantized, and if they are quantized, the position coordinates obtained by the hybrid decoding are inversely quantized. In this embodiment, the code stream (for example, the quantization flag in the code stream) is used to determine whether the point cloud data is quantized during encoding. If quantization is performed, the position in the mixed decoding result obtained in step S520 needs to be determined. The decoding result is inversely quantized to obtain the position coordinates of the point cloud data.

Further, the decoding method 500 may further include (not shown in FIG. 5): determining the point cloud data after the position coordinates of the point cloud data are quantified in the encoding process of the point cloud data In the encoding process, whether the attribute values of the point cloud data with the same position coordinates obtained by the quantization are merged; if it is determined that the merge is performed, the position decoding and attribute decoding of the point cloud data of any leaf node includes: Decoding the unique attribute value of the point cloud data in any leaf node; if it is determined that no merging has been performed, performing position decoding and attribute decoding on the point cloud data of any leaf node includes: decoding the point in any leaf node The number of cloud data and the attribute value are decoded separately.

In this embodiment, it has been determined that the position coordinates of the point cloud data are quantized during the encoding process. Since the position coordinates of all the point cloud data are quantized, there may be repeated coordinate values. At this time, the same position Coordinates correspond to multiple point cloud data, that is, the same location coordinate may correspond to multiple attribute values. Therefore, it can be further determined whether the attribute values of the point cloud data with the same position coordinates obtained by quantization are merged (for example, determined by the attribute merge flag bit, of course, it can also be merged or not merged by default without the flag bit and the Confirm the steps). If it is determined that the attribute values of the point cloud data with the same position coordinates obtained after quantization are combined, it means that each leaf node can be regarded as including only one point cloud data point, and the attribute value is also one. Therefore, you can directly The unique attribute value of the point cloud data in each leaf node is decoded, without the need to decode the number of point cloud data, because the number of point cloud data in each leaf node is one, and there is no need to waste encoding bit pairs during encoding. It is encoded. Conversely, if it is determined that the attribute values of the point cloud data with the same position coordinates obtained after quantization are not merged, it means that the number of point cloud data and the attribute value of each point cloud data need to be decoded when decoding the leaf nodes.

In another embodiment of the present invention, the method 500 may further include (not shown in FIG. 5): determining, based on the code stream, whether the point cloud data is encoded in the process of encoding more than one point cloud data. The attribute values of the point cloud data in any leaf node are merged; if it is determined that the merging is carried out, performing position decoding and attribute decoding on the point cloud data of any leaf node includes: Decoding the merged attribute value of the point cloud data of the point cloud; if it is determined that the merge is not performed, the position decoding and attribute decoding of the point cloud data of any leaf node includes: The number and attribute value are decoded separately.

In this embodiment, it can be determined based on the code stream whether the attribute values of the point cloud data in the leaf nodes that include more than one point cloud data are merged in the encoding process (for example, determined by the attribute merge flag in the code stream, Of course, it can also be merged or not merged by default without the flag bit and the determination step). If it is determined that the attribute values of the point cloud data in the leaf nodes that include more than one point cloud data are merged, it means that when decoding For such leaf nodes, only the unique attribute value needs to be decoded. Conversely, if it is determined that the attribute values of the point cloud data in the leaf nodes that include more than one point cloud data are not merged, it means that the number of point cloud data and the attributes of each point cloud data need to be decoded for such leaf nodes during decoding value.

The above exemplarily describes the point cloud data decoding method according to the embodiment of the present invention. Based on the above description, the point cloud data decoding method according to the embodiment of the present invention adopts a decoding scheme that combines position coordinate decoding and attribute decoding, so that when part of the point cloud data needs to be used, there is no need to decode the position coordinates of all the point cloud data. Then decode the attribute value to improve the decoding efficiency. Exemplarily, the method for decoding point cloud data according to an embodiment of the present invention may be implemented in a device, device, or system having a memory and a processor.

The following describes a point cloud data decoding system according to an embodiment of the present invention with reference to FIG. 6. Fig. 6 shows a schematic block diagram of a point cloud data decoding system 600 according to an embodiment of the present invention. The point cloud data decoding system 600 includes a storage device 610 and a processor 620.

Wherein, the storage device 610 stores a program for implementing the corresponding steps in the point cloud data decoding method according to the embodiment of the present invention. The processor 620 is configured to run a program stored in the storage device 610 to execute corresponding steps of the point cloud data decoding method according to the embodiment of the present invention.

In one embodiment, when the program is run by the processor 620, the point cloud data decoding system 600 is caused to perform the following steps: perform arithmetic decoding on the bitstream of the point cloud data to obtain the arithmetic decoding result; and perform the arithmetic decoding result on the arithmetic decoding result Performing hybrid decoding of position coordinates and attributes to obtain respective position coordinates and attribute values of the point cloud data; wherein, the hybrid decoding includes: performing position coordinate decoding based on multi-tree division to obtain multiple leaf nodes, one of which is The attribute data of the leaf node is located between the position data of the one leaf node and the position data of another leaf node in the code stream; the position decoding and attribute decoding are performed on the point cloud data contained in the leaf node.

In an embodiment of the present invention, when the program is run by the processor 620, the point cloud data decoding system 600 is also caused to perform the following steps: based on the code stream, it is determined whether the point cloud data is correctly encoded during the encoding process. The position coordinates of the point cloud data are quantized, and if quantized, the position coordinates obtained by the hybrid decoding are inversely quantized.

In an embodiment of the present invention, when the program is run by the processor 620, the point cloud data decoding system 600 is also caused to perform the following steps: in the process of determining the encoding of the point cloud data, After the position coordinates are quantified, it is determined whether the attribute values of the quantized point cloud data with the same position coordinates are merged in the encoding process of the point cloud data; if it is determined that the merge is performed, the point of any leaf node is determined The position decoding and attribute decoding of cloud data includes: decoding the unique attribute value of the point cloud data in any leaf node; if it is determined that no merging is performed, then performing position decoding and attribute decoding on the point cloud data of any leaf node The decoding includes: respectively decoding the number and attribute value of the point cloud data in any of the leaf nodes.

In an embodiment of the present invention, when the program is run by the processor 620, the point cloud data decoding system 600 is also caused to perform the following steps: based on the code stream, it is determined whether the point cloud data is correctly encoded during the encoding process. The attribute values of the point cloud data in any leaf node that includes more than one point cloud data are merged; if it is determined that the merge is performed, the position decoding and attribute decoding on the point cloud data of any leaf node includes: Decoding the merged attribute value of the point cloud data in any one of the leaf nodes; if it is determined that the merge has not been performed, then performing position decoding and attribute decoding on the point cloud data of any one of the leaf nodes includes: The number and attribute value of the point cloud data in the node are decoded separately.

In an embodiment of the present invention, when the program is run by the processor 620, the point cloud data decoding system 600 is also caused to perform the following steps: determine whether to correct in the encoding process based on the attribute merging flag bit in the code stream The attribute values of the point cloud data are merged, and the position decoding and attribute decoding are performed on the leaf node based on the result of the determination.

In an embodiment of the present invention, when the program is executed by the processor 620, the decoding system 600 for point cloud data is executed to decode the attribute value of the point cloud data, including: The attribute value is decoded into binarization.

In an embodiment of the present invention, when the program is executed by the processor 620, the point cloud data decoding system 600 executes the binarization decoding of the attribute value of the point cloud data, including: The attribute values of the point cloud data are subjected to fixed-length decoding, truncated Rice decoding, or exponential Columbus decoding.

In addition, according to an embodiment of the present invention, there is also provided a storage medium on which program instructions are stored, and the program instructions are used to execute the point cloud data in the embodiments of the present invention when the program instructions are run by a computer or a processor. The corresponding steps of the decoding method. The storage medium may include, for example, a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

In an embodiment, the computer program instructions can execute the point cloud data decoding method according to the embodiment of the present invention when being executed by a computer.

In one embodiment, the computer program instructions, when run by the computer or processor, cause the computer or processor to perform the following steps: perform arithmetic decoding on the code stream of the point cloud data to obtain the arithmetic decoding result; and perform the arithmetic decoding result on the arithmetic decoding result Performing hybrid decoding of position coordinates and attributes to obtain respective position coordinates and attribute values of the point cloud data; wherein, the hybrid decoding includes: performing position coordinate decoding based on multi-tree division to obtain multiple leaf nodes, one of which is The attribute data of the leaf node is located between the position data of the one leaf node and the position data of another leaf node in the code stream; the position decoding and attribute decoding are performed on the point cloud data contained in the leaf node.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor also executes the following steps: based on the code stream, it is determined whether the point cloud data is correct or not during the encoding process of the point cloud data. The position coordinates of the point cloud data are quantized, and if quantized, the position coordinates obtained by the hybrid decoding are inversely quantized.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor also executes the following steps: in the process of determining the encoding of the point cloud data, After the position coordinates are quantified, it is determined whether the attribute values of the quantized point cloud data with the same position coordinates are merged in the encoding process of the point cloud data; if it is determined that the merge is performed, the point of any leaf node is determined The position decoding and attribute decoding of cloud data includes: decoding the unique attribute value of the point cloud data in any leaf node; if it is determined that no merging is performed, then performing position decoding and attribute decoding on the point cloud data of any leaf node The decoding includes: respectively decoding the number and attribute value of the point cloud data in any of the leaf nodes.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or the processor, the computer or the processor also executes the following steps: based on the code stream, it is determined whether the point cloud data is correct or not during the encoding process of the point cloud data. The attribute values of the point cloud data in any leaf node that includes more than one point cloud data are merged; if it is determined that the merge is performed, the position decoding and attribute decoding on the point cloud data of any leaf node includes: Decoding the merged attribute value of the point cloud data in any one of the leaf nodes; if it is determined that the merge has not been performed, then performing position decoding and attribute decoding on the point cloud data of any one of the leaf nodes includes: The number and attribute value of the point cloud data in the node are decoded separately.

In an embodiment of the present invention, when the computer program instructions are run by the computer or the processor, the computer or the processor also executes the following steps: based on the attribute merging flag bit in the code stream, it is determined whether to correct during the encoding process. The attribute values of the point cloud data are merged, and the position decoding and attribute decoding are performed on the leaf node based on the result of the determination.

In an embodiment of the present invention, when the computer program instructions are executed by the computer or processor to decode the attribute values of the point cloud data, the computer program instructions include: The attribute value is decoded into binarization.

In an embodiment of the present invention, when the computer program instructions cause the computer or the processor to execute the binary decoding of the attribute value of the point cloud data when being executed by the computer or the processor, the method includes: The attribute values of the point cloud data are subjected to fixed-length decoding, truncated Rice decoding, or exponential Columbus decoding.

The foregoing exemplarily describes the point cloud data encoding and decoding method and system according to the embodiments of the present invention, and the point cloud data collection device for the storage medium. Based on the above description, the point cloud data coding and decoding method, system and storage medium according to the embodiments of the present invention combine the characteristics of position coordinate coding and the characteristics of attribute coding, and adopt a coding scheme that combines position coordinate coding and attribute coding, so that When it is necessary to use part of the point cloud data, it is not necessary to decode the position coordinates of all the point cloud data and then decode the attribute value, thereby improving the decoding efficiency.

Although the exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above-described exemplary embodiments are merely exemplary, and are not intended to limit the scope of the present invention thereto. Those of ordinary skill in the art can make various changes and modifications therein without departing from the scope and spirit of the present invention. All these changes and modifications are intended to be included within the scope of the present invention as claimed in the appended claims.

A person of ordinary skill in the art may be aware that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered as going beyond the scope of the present invention.

In the several embodiments provided in this application, it should be understood that the disclosed device and method may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or It can be integrated into another device, or some features can be ignored or not implemented.

In the instructions provided here, a lot of specific details are explained. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and technologies are not shown in detail, so as not to obscure the understanding of this specification.

Similarly, it should be understood that in order to simplify the present invention and help understand one or more of the various aspects of the invention, in the description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment. , Or in its description. However, the method of the present invention should not be construed as reflecting the intention that the claimed invention requires more features than those explicitly stated in each claim. To be more precise, as reflected in the corresponding claims, the point of the invention is that the corresponding technical problems can be solved with features that are less than all the features of a single disclosed embodiment. Therefore, the claims following the specific embodiment are thus explicitly incorporated into the specific embodiment, wherein each claim itself serves as a separate embodiment of the present invention.

Those skilled in the art can understand that in addition to mutual exclusion between the features, any combination of all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and any method or device disclosed in this manner can be used. Processes or units are combined. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

In addition, those skilled in the art can understand that although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments means that they are within the scope of the present invention. Within and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

The various component embodiments of the present invention may be implemented by hardware, or by software modules running on one or more processors, or by a combination of them. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the functions of some modules according to the embodiments of the present invention. The present invention can also be implemented as a device program (for example, a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for realizing the present invention may be stored on a computer-readable medium, or may have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be constructed as a limitation to the claims. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims that list several devices, several of these devices may be embodied in the same hardware item. The use of the words first, second, and third, etc. do not indicate any order. These words can be interpreted as names.

The above are only specific implementations or descriptions of specific implementations of the present invention. The protection scope of the present invention is not limited thereto. Any person skilled in the art can easily fall within the technical scope disclosed by the present invention. Any change or replacement should be included in the protection scope of the present invention. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (20)

A coding method for point cloud data, characterized in that the coding method includes: Perform hybrid coding of position coordinates and attributes on the input point cloud data; and Performing arithmetic coding on the result of the hybrid coding to obtain the code stream of the point cloud data; Wherein, the hybrid encoding includes: performing multi-tree division on the space where the point cloud data is located to obtain multiple leaf nodes, and performing position encoding and attribute encoding on the point cloud data in the leaf nodes; among them, one leaf node The attribute data is located between the position data of the one leaf node and the position data of another leaf node in the code stream. The coding method according to claim 1, wherein the coding method further comprises: Before performing the hybrid encoding of position coordinates and attributes on the input point cloud data, quantize the position coordinates of the input point cloud data, and perform the hybrid encoding based on the quantized position coordinates. The coding method according to claim 2, wherein the coding method further comprises: After quantifying the position coordinates of the input point cloud data, merge the attribute values of the point cloud data obtained by quantizing the same position coordinates; and Performing position coding and attribute coding on the point cloud data of any leaf node includes: coding the unique attribute value of the point cloud data in any leaf node. The coding method according to claim 2, wherein the coding method further comprises: After the position coordinates of the input point cloud data are quantified, the attribute values of the point cloud data that have the same position coordinates obtained by quantization are not merged; and Performing position coding and attribute coding on the point cloud data of any leaf node includes: respectively coding the number and attribute values of the point cloud data in any leaf node. The coding method according to claim 1, wherein the coding method further comprises: When any leaf node includes more than one point cloud data, the attribute values of the point cloud data in any leaf node are merged; and Performing position coding and attribute coding on the point cloud data of any leaf node includes: coding a merged attribute value of the point cloud data in any leaf node. The coding method according to claim 1, wherein the coding method further comprises: When any leaf node includes more than one point cloud data, the attribute values of the point cloud data in any leaf node are not merged; and Performing position coding and attribute coding on the point cloud data of any leaf node includes: respectively coding the number and attribute values of the point cloud data in any leaf node. The encoding method according to any one of claims 3-6, wherein the code stream includes an attribute merging flag bit, and the value of the attribute merging flag bit indicates whether the point cloud is adjusted during the encoding process. The attribute values of the data are merged. The encoding method according to any one of claims 1-7, wherein encoding the attribute value of the point cloud data comprises: Binary encoding is performed on the attribute value of the point cloud data. The coding method according to claim 8, wherein the binarization coding of the attribute value of the point cloud data comprises: Perform fixed-length coding, truncated Rice coding, or exponential Golomb coding on the attribute value of the point cloud data. A method for decoding point cloud data, characterized in that the decoding method includes: Perform arithmetic decoding on the bitstream of the point cloud data to obtain the arithmetic decoding result; and Performing hybrid decoding of position coordinates and attributes on the arithmetic decoding result to obtain respective position coordinates and attribute values of the point cloud data; Wherein, the hybrid decoding includes: decoding the position coordinates based on the multi-tree division to obtain a plurality of leaf nodes, wherein the attribute data of one leaf node is located in the code stream between the position data of the one leaf node and the position data of the other leaf node. Between the position data of the nodes; perform position decoding and attribute decoding on the point cloud data contained in the leaf nodes. The decoding method according to claim 10, wherein the decoding method further comprises: Based on the code stream, it is determined whether the position coordinates of the point cloud data are quantized during the encoding process of the point cloud data, and if quantization is performed, the position coordinates obtained by the hybrid decoding are inversely quantized. The decoding method according to claim 11, wherein the decoding method further comprises: After the position coordinates of the point cloud data are quantified in the encoding process of the point cloud data, it is determined whether the point cloud data is quantized to obtain the attribute values of the point cloud data with the same position coordinates in the encoding process. Merged If it is determined that the merging is performed, performing position decoding and attribute decoding on the point cloud data of any leaf node includes: decoding the unique attribute value of the point cloud data in any leaf node; If it is determined that no merging is performed, performing position decoding and attribute decoding on the point cloud data of any leaf node includes: decoding the number and attribute values of the point cloud data in any leaf node respectively. The decoding method according to claim 10, wherein the decoding method further comprises: Determining, based on the code stream, whether attribute values of the point cloud data in any leaf node including more than one point cloud data are merged during the encoding process of the point cloud data; If it is determined that the merging is performed, performing position decoding and attribute decoding on the point cloud data of any leaf node includes: decoding the merged attribute value of the point cloud data in any leaf node; If it is determined that no merging is performed, performing position decoding and attribute decoding on the point cloud data of any leaf node includes: decoding the number and attribute values of the point cloud data in any leaf node respectively. The decoding method according to any one of claims 10 to 13, wherein the decoding method further comprises: Determine whether the attribute values of the point cloud data are merged during the encoding process based on the attribute merging flag bit in the code stream, and perform position decoding and attribute decoding on the leaf node based on the result of the determination. The decoding method according to any one of claims 10-14, wherein decoding the attribute value of the point cloud data comprises: Binary decoding is performed on the attribute value of the point cloud data. The decoding method according to claim 15, wherein the binarization decoding of the attribute value of the point cloud data comprises: Perform fixed-length decoding, truncated Rice decoding, or exponential Golomb decoding on the attribute value of the point cloud data. A point cloud data coding system, characterized in that the coding system includes a storage device and a processor, the storage device stores a computer program run by the processor, and the computer program is being processed by the processor. The point cloud data encoding method according to any one of claims 1-9 is executed when the device is running. A point cloud data decoding system, characterized in that the decoding system includes a storage device and a processor, the storage device stores a computer program run by the processor, and the computer program is being processed by the processor. The point cloud data decoding method according to any one of claims 10-16 is executed when the device is running. A storage medium, characterized in that a computer program is stored on the storage medium, and the computer program executes the point cloud data encoding method according to any one of claims 1-9 when the computer program is running. A storage medium, characterized in that a computer program is stored on the storage medium, and the computer program executes the point cloud data decoding method according to any one of claims 10-16 when the computer program is running.

Images