CN113473153A - Point cloud attribute prediction method, encoding method, decoding method and equipment thereof - Google Patents
Point cloud attribute prediction method, encoding method, decoding method and equipment thereof Download PDFInfo
- Publication number
- CN113473153A CN113473153A CN202010239384.8A CN202010239384A CN113473153A CN 113473153 A CN113473153 A CN 113473153A CN 202010239384 A CN202010239384 A CN 202010239384A CN 113473153 A CN113473153 A CN 113473153A
- Authority
- CN
- China
- Prior art keywords
- attribute
- point
- prediction
- current node
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/593—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/13—Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/96—Tree coding, e.g. quad-tree coding
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a point cloud attribute prediction method, an encoding method, a decoding method and equipment thereof, wherein the point cloud attribute prediction method comprises the following steps: adding an offset value to the original point cloud coordinate to obtain a new coordinate value; determining an offset morton sequence according to the new coordinate values; and determining the attribute predicted value of the current node according to the offset morton sequence. According to the method, a neighbor point closest to a current node in a physical space is found by adopting an offset Morton sequence, so that an attribute predicted value of the current node is determined, and finally, attribute coding is carried out according to the attribute predicted value or an attribute value of the current node is determined according to the attribute predicted value of the current node and a point cloud attribute code stream. The point cloud attribute prediction method, the point cloud attribute encoding method and the point cloud attribute decoding method can improve the utilization of the geometrical information and attribute information correlation of the point cloud, so that the compression performance of the point cloud attributes is improved.
Description
Technical Field
The invention relates to the technical field of point cloud processing, in particular to a point cloud attribute prediction method, an encoding method, a decoding method and equipment thereof.
Background
Three-dimensional point clouds are an important manifestation of real-world digitization. With the rapid development of three-dimensional scanning devices (e.g., lasers, radars, etc.), the accuracy and resolution of the point cloud becomes higher. The high-precision point cloud is widely applied to the construction of urban digital maps and plays a technical support role in numerous popular researches such as smart cities, unmanned driving, cultural relic protection and the like. The point cloud is obtained by sampling the surface of an object by a three-dimensional scanning device, the number of points of one frame of point cloud is generally in the million level, each point contains geometric information and attribute information such as color and reflectivity, and the data volume is huge. The huge data volume of the three-dimensional point cloud brings huge challenges to data storage, transmission and the like, so that the point cloud compression becomes very important.
Point cloud compression is mainly divided into geometric compression and attribute compression, and the point cloud attribute compression method described in a test platform PCEM provided by the AVS (Audio Video coding Standard) point cloud compression working group of China at present mainly adopts a point cloud prediction method based on a morton sequence, namely, morton sorting is carried out on current node cloud according to position information of the point cloud, an attribute value of a previous point in the morton sequence of the current node is selected as an attribute predicted value of the current node, and finally, the actual attribute value of the current node is subtracted from the attribute predicted value to obtain an attribute residual value.
However, the point cloud prediction method only considers the morton sequence, and there is a case that the attribute value of the current node cannot be well predicted by the previous point of the morton sequence, which easily causes low accuracy of attribute prediction, thereby reducing the encoding and decoding performance.
Accordingly, the prior art is yet to be improved and developed.
Disclosure of Invention
The invention provides a point cloud attribute prediction method, an encoding method, a decoding method and equipment thereof, aiming at solving the problem that the point cloud attribute encoding and decoding performance is poor due to the fact that neighbors found by point clouds in attribute encoding are not close enough to influence an attribute predicted value in the prior art.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
a point cloud attribute prediction method comprises the following steps:
adding an offset value to the original point cloud coordinate to obtain a new coordinate value;
determining an offset morton sequence according to the new coordinate values;
and determining the attribute predicted value of the current node according to the offset morton sequence.
The point cloud attribute prediction method, wherein the determining of the attribute prediction value of the current node according to the offset morton sequence, comprises the following steps:
searching a first existing node forward according to the offset morton sequence to serve as a first prediction point;
taking the attribute value of the first prediction point as an attribute prediction value of the current node;
or, searching K1 existing nodes forward according to the offset morton sequence to serve as first prediction points, and finding the point with the minimum distance from the current node at the K1 first prediction points;
taking the attribute value of the point with the minimum distance as an attribute predicted value of the current node;
or, searching K1 existing nodes forward according to the offset morton sequence to serve as first prediction points, and weighting the attribute values of the K1 first prediction points to serve as the attribute prediction values of the current node.
The point cloud attribute prediction method, wherein the determining of the attribute prediction value of the current node according to the offset morton sequence, comprises the following steps:
searching a first existing node forward according to the offset morton sequence to serve as a first prediction point;
searching a first existing node forward according to the original Morton sequence under the point cloud original coordinate to serve as a second prediction point;
calculating the distance d1 from the first predicted point to the current node;
calculating the distance d2 from the second predicted point to the current node;
comparing the distance d1 with the distance d2, and selecting a point with a small distance as a third predicted point;
taking the attribute value of the third prediction point as the attribute prediction value of the current node;
searching K1 existing nodes forward according to the offset Morton sequence as first prediction points; according to the point
K2 existing nodes are searched forward in the original morton sequence under the cloud original coordinate to serve as second prediction points;
calculating the distances from the first predicted point and the second predicted point to the current node, and selecting one or more points with small distances as third predicted points;
taking the attribute value or the attribute weighted value of the third predicted point as the attribute predicted value of the current node;
or, searching K1 existing nodes forward according to the offset Morton sequence as first prediction points;
searching K2 existing nodes forward according to the original Morton sequence under the point cloud original coordinate to serve as second prediction points;
calculating a weighted distance d1 from the first predicted point to the current node;
calculating a weighted distance d2 from the second predicted point to the current node;
comparing the distance d1 with the distance d2, and selecting a point with a small distance as a third predicted point;
and taking the attribute weighted value of the third prediction point as the attribute prediction value of the current node.
The point cloud attribute prediction method, wherein the step of calculating the distance from the first predicted point to the current node comprises:
calculating the Euclidean distance between the current node and the first prediction point as the distance between the current node and the first prediction point;
or calculating the maximum value of the absolute value of the coordinate difference between the current node and the first prediction point in the X, Y and Z directions as the distance between the current node and the first prediction point;
or, calculating the sum of the difference values of the current node and the first prediction point coordinate in the X, Y and Z directions as the distance between the current node and the first prediction point.
A point cloud attribute prediction device comprises a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor;
the communication bus realizes connection communication between the processor and the memory;
the processor, when executing the computer readable program, implements the steps in the point cloud attribute prediction method of the present invention.
A point cloud attribute coding method comprises the following steps:
determining an attribute predicted value of a current node by adopting the point cloud attribute prediction method;
determining a prediction residual value of the current node according to a difference value between the attribute value of the current node and the attribute predicted value of the current node;
and coding the prediction residual value of the current node to obtain a point cloud code stream.
A point cloud attribute coding device comprises a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor;
the communication bus realizes connection communication between the processor and the memory;
the processor, when executing the computer readable program, implements the steps in the point cloud attribute encoding method according to the present invention.
A point cloud attribute decoding method comprises the following steps:
decoding the point cloud attribute code stream to obtain an attribute residual value of the current node;
determining an attribute predicted value of a current node by adopting the point cloud attribute prediction method;
and determining the attribute value of the current node according to the sum of the attribute predicted value of the current node and the attribute residual value.
A point cloud attribute decoding device comprises a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor;
the communication bus realizes connection communication between the processor and the memory;
the processor, when executing the computer readable program, implements the steps in the point cloud attribute decoding method of the present invention.
Has the advantages that: compared with the prior art, the method and the device have the advantages that the nearest neighbor point in the physical space of the current node is found by adopting the offset Morton sequence, so that the attribute predicted value of the current node is determined, and finally, the attribute coding is carried out according to the attribute predicted value or the attribute value of the current node is determined according to the attribute predicted value of the current node and the point cloud attribute code stream. The point cloud attribute prediction method, the point cloud attribute encoding method and the point cloud attribute decoding method can improve the utilization of the geometrical information and attribute information correlation of the point cloud, so that the compression performance of the point cloud attributes is improved.
Drawings
Fig. 1 is a schematic diagram of an original morton sequence of nodes in an octree peer-to-peer cloud according to an embodiment of the present invention.
Fig. 2 is a flowchart illustrating a point cloud attribute prediction method according to a preferred embodiment of the present invention.
FIG. 3 is a schematic diagram of an offset morton sequence of nodes in the point cloud under the octree according to the embodiment of the present invention.
Fig. 4 is a schematic diagram of an original morton sequence of nodes in a peer cloud under a quad tree according to an embodiment of the present invention.
FIG. 5 is a schematic diagram of an offset morton sequence of nodes in a point cloud under a quad tree according to an embodiment of the present invention.
Fig. 6 is a flowchart of a point cloud attribute encoding method according to a preferred embodiment of the present invention.
Fig. 7 is a schematic structural diagram of a point cloud attribute encoding apparatus according to the present invention.
Fig. 8 is a flowchart illustrating a point cloud attribute decoding method according to a preferred embodiment of the present invention.
Detailed Description
The invention provides a point cloud attribute prediction method, an encoding method, a decoding method and equipment thereof, and in order to make the purpose, technical scheme and effect of the invention clearer and clearer, the invention is further described in detail by referring to the attached drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The invention will be further explained by the description of the embodiments with reference to the drawings.
Specifically, the morton sequence after morton sorting of the points in the point cloud can be specifically expressed as: the position coordinates (Xk, Yk, Zk) of the k-th point are expressed as:
the morton code for the k point is expressed as:
or representing every third bit by an octal number
The morton code for the kth point can be expressed as:
FIG. 1 is a partial schematic view of a three-dimensional point cloud, with numbers representing the Morton order in the three-dimensional point cloud, assuming that points exist at positions in the Morton order of 0, 2, 10, 16, the three points being numbered with letters A, B, C, D for ease of presentation. In the attribute coding of the PCEM coder, according to the searching mode of the point cloud prediction points in the morton sequence, searching the first 1 point in the morton sequence as the current prediction point. Assuming that the current node is D (16), according to the search mode of PCEM, the first 1 point C (10) in the morton sequence is found as the predicted point of D (16), so that a problem exists in that B (2) is a point closer to D (16) but the predicted point of D (16) is a C (10) farther away, and the attribute value of C (10) is used for predicting D (16) to influence the point cloud compression performance.
Based on the problems in the above technology, the present invention provides a flow chart of a preferred embodiment of a point cloud attribute prediction method, as shown in fig. 2, which includes the steps of:
s1, adding an offset value to the original point cloud coordinate to obtain a new coordinate value;
s2, determining the offset Morton sequence according to the new coordinate value;
and S3, determining the attribute predicted value of the current node according to the offset Morton sequence.
Specifically, in this embodiment, a fixed offset value (j1, j2, j3) is added to the original coordinates (x, y, z) of the point cloud, a morton code corresponding to the point cloud is generated by using the new coordinates (x + j1, y + j2, z + j3), an offset morton sequence is obtained according to the morton sorting, and the values of the offset values j1, j2, j3 are greater than or equal to 1. In this embodiment, the offset morton sequence is adopted to find the nearest neighbor point to the current node physically, so as to determine the attribute predicted value of the current node, and finally, the attribute coding is performed according to the attribute predicted value or the attribute value of the current node is determined according to the attribute predicted value of the current node and the point cloud attribute code stream, so as to improve the coding and decoding performance of the point cloud attribute.
In some specific embodiments, offset values (1, 1, 1) are added to the 0, 2, 10, 16 points in the origin cloud to obtain 7, 21, 29, 23 points in the new point cloud after offset, as shown in fig. 3, that is, A, B, C, D corresponds to the points numbered 0, 2, 10, 16 in the original morton sequence and corresponds to the points numbered 7, 21, 29, 23 in the new morton sequence, and this new morton sequence is called offset morton sequence.
In some embodiments, the first existing node is looked forward as the first predicted point according to the offset morton sequence; and taking the attribute value of the first prediction point as an attribute prediction value of the current node.
As an example, the first existing point is found forward in offset morton order as the predicted point, and in offset morton order as shown in fig. 3, the current node D (23) finds the first 1 point in offset morton order as the predicted point, and the found predicted point is B (21), and the predicted point of B (21) is used as the predicted point of the current node D (23).
In some embodiments, K1 existing nodes are searched forward according to the offset morton sequence as first prediction points, and the point with the minimum distance from the current node is found at the K1 first prediction points; and taking the attribute value of the point with the minimum distance as the attribute predicted value of the current node.
For example, assuming that K1 is set to 2, in the offset morton sequence shown in fig. 3, the current node D (23) searches the first 2 points in the offset morton sequence as the predicted points, the found predicted points are B (21) and a (7), the first-order distance between the current nodes D (23) and B (21) is 1, the first-order distance between the current nodes D (23) and a (7) is 2, and B (21) with a smaller distance is selected as the predicted point of the current node D (23).
In some embodiments, K1 existing nodes are looked forward according to the offset morton order as first prediction points, and the attribute values of the K1 first prediction points are weighted as the attribute prediction values of the current node.
For example, assuming that K1 is set to 2, in the offset morton order shown in fig. 3, the current node D (23) finds the first 2 points in the offset morton order as predicted points, the found predicted points are B (21) and a (7), and the average of the attribute values of B (21) and a (7) is used as the predicted value of the current node D (23).
In some embodiments, the first existing node is looked forward as the first predicted point according to the offset morton sequence; searching a first existing node forward according to the original Morton sequence under the point cloud original coordinate to serve as a second prediction point; calculating the distance d1 from the first predicted point to the current node; calculating the distance d2 from the second predicted point to the current node; comparing the distance d1 with the distance d2, and selecting a point with a small distance as a third predicted point; and taking the attribute value of the third prediction point as the attribute prediction value of the current node.
By way of example, the first existing node of current node D (16) is found in the original Morton order as shown in FIG. 1, the point found is C (10), C (10) is taken as predicted point 1, and the first order distance is 3. The first existing point of the current node D (23) is found in the offset morton sequence as shown in fig. 3, the first 1 point in the morton sequence is found by the current node D (23) as the predicted point, the found predicted point is B (21), B (21) is taken as the predicted point 2, and the first-order distance is 1. Comparing the distances from the prediction point 1 and the prediction point 2 to the current node, selecting the prediction point 2 with a smaller distance as the prediction point of the current node D (23), and performing attribute prediction on the current node by using the attribute value of the prediction point 2.
In some embodiments, K1 existing nodes are looked forward as first predicted points according to the offset morton order; searching K2 existing nodes forward according to the original Morton sequence under the point cloud original coordinate to serve as second prediction points; calculating the distances from the first prediction point and the second prediction point to the current node, and selecting a point with a small distance as a third prediction point; and taking the attribute value of the third prediction point as the attribute prediction value of the current node.
By way of example, K1 existing points are found forward in the offset morton sequence shown in fig. 3, K2 existing points are found forward in the original morton sequence shown in fig. 1, and the point with the smallest distance from the current point is selected from K1+ K2 as the predicted point. In this example, K1 is set to 2, K2 is set to 2, and in the original morton sequence, the current point D (16) searches the first 2 points in the original morton sequence as prediction points, and the prediction points found are C (10) and B (2). In the offset morton sequence, the current point D (23) searches the first 2 points in the offset morton sequence as prediction points, and the prediction points found are B (21) and a (7). The collection of points found in the two morton sequences is A, B, C, the first-order distances from the current point D are 2, 1 and 4 respectively, and B with the smallest distance is selected as the predicted point of the current point D.
In some embodiments, K1 existing nodes are looked forward as first predicted points according to the offset morton order; searching K2 existing nodes forward according to the original Morton sequence under the point cloud original coordinate to serve as second prediction points; calculating a weighted distance d1 from the first predicted point to the current node; calculating a weighted distance d2 from the second predicted point to the current node; comparing the distance d1 with the distance d2, and selecting a point with a small distance as a third predicted point; and taking the attribute weighted value of the third prediction point as the attribute prediction value of the current node.
For example, K1 existing points are found forward in the offset morton sequence shown in fig. 3, K2 existing points are found forward in the original morton sequence shown in fig. 1, the weighted distance from the K1 point to the current point and the weighted distance from the K2 point to the current point are respectively calculated, K1 points or K2 points with smaller distances are selected as predicted points, and attribute weighted values of the predicted points are used as predicted values. In this example K1 is set to 2, K2 is set to 2, and the distance weighted manner is the mean. In the original morton sequence, the current point D (16) searches the first 2 points in the original morton sequence as prediction points, the found prediction points are C (10) and B (2), the first-order distances from the current point D (16) are 4 and 1, and the average value of the distances is 2.5. In the offset morton sequence, the current point D (23) searches the first 2 points in the offset morton sequence as prediction points, the first-order distances between B (21) and A (7) and the current point D (16) are 1 and 2, and the average value of the distances is 1.5. B (21) and A (7) in the offset Morton order with a smaller weighting distance are selected as prediction points, and the attribute weighting values of B (21) and A (7) are selected as prediction values.
In some embodiments, K1 existing points are found forward in offset morton order, K2 existing points are found forward in original morton order, the closest KX point among the K1+ K2 points is selected as the predicted point, and KX attribute weighted values are used as the predicted values. In this example K1 is set to 2, K2 is set to 2, KX is set to 2, and the distance weighting is a mean. In the original morton sequence, the current point D (23) searches the first 2 points in the original morton sequence as prediction points, and the found prediction points are C (10) and B (2). In the offset morton sequence, the current point D (23) searches the first 2 points in the offset morton sequence as prediction points, and the prediction points found are B (21) and a (7). The collection of points found in the two morton sequences is A, B, C, the first-order distances from the current point D are 2, 1 and 4 respectively, 2 points A, B with smaller first-order distances are selected as the prediction points of the current point D, and the attribute weighted value of A, B is used as the prediction value.
In some embodiments, the step of calculating the distance of the first predicted point to the current node comprises: calculating the Euclidean distance between the current node and the first prediction point as the distance between the current node and the first prediction point; or calculating the maximum value of the absolute value of the coordinate difference between the current node and the first prediction point in the X, Y and Z directions as the distance between the current node and the first prediction point; or, calculating the sum of the difference values of the current node and the first prediction point coordinate in the X, Y and Z directions as the distance between the current node and the first prediction point.
In some embodiments, in the quad-tree case, the point cloud has a total of 5 points in the original morton sequence as shown in fig. 4, morton numbers 8, 14, 19, 24, 25, respectively, which are referred to for convenience of expression as A, B, C, D, E. In the original morton sequence, the previous morton sequence prediction point found by the current point E (25) is D (24), the first-order distance between D (24) and the current point E (25) is 4, and D (24) is the first prediction point. Adding offset (1, 1) to the coordinates of the points in fig. 4 to obtain the points in fig. 5, forming an offset morton sequence, A, B, C, D, E finding the predicted point of the current point E (28) from the offset morton sequence, finding B (27) and C (26) in the offset morton sequence, wherein the previous morton sequence predicted point found by a (25) is B (27), the first-order distance from the current point is 1, and B (27) is taken as the second predicted point, and the predicted point is numbered 25, 27, 26, 30, 28 corresponding to the offset morton sequence in fig. 5. According to the principle of finding the nearest prediction point, B (27) which is the nearest to the prediction point of the current point is selected to be closer than the prediction point D (24) found in the original Morton sequence, so that the point B is selected to be the prediction point of the current point E.
In some embodiments, a point cloud attribute prediction device is also provided, wherein the point cloud attribute prediction device comprises a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor; the communication bus realizes connection communication between the processor and the memory; the processor, when executing the computer readable program, implements the steps in the point cloud attribute prediction method of the present invention.
In some embodiments, there is also provided a point cloud attribute encoding method, as shown in fig. 6, which includes the steps of:
s10, determining an attribute predicted value of the current node by adopting the point cloud attribute prediction method;
s20, determining a prediction residual value of the current node according to a difference value between the attribute value of the current node and the attribute prediction value of the current node;
and S30, coding the prediction residual value of the current node to obtain a point cloud code stream.
In this embodiment, a neighbor point that is physically closest to the current node is found by using an offset morton sequence, so as to determine an attribute prediction value of the current node, and finally, attribute encoding is performed according to the attribute prediction value. The point cloud attribute coding method provided by the embodiment can improve the utilization of the geometrical information and attribute information correlation of the point cloud, thereby improving the coding and decoding performance of the point cloud attribute.
Based on the point cloud attribute encoding method, the present invention further provides a point cloud attribute encoding apparatus, wherein, as shown in fig. 7, the apparatus comprises at least one processor (processor) 20; a display screen 21; and a memory (memory)22, and may further include a communication Interface (Communications Interface)23 and a bus 24. The processor 20, the display 21, the memory 22 and the communication interface 23 can communicate with each other through the bus 24. The display screen 21 is configured to display a user guidance interface preset in the initial setting mode. The communication interface 23 may transmit information. The processor 20 may call logic instructions in the memory 22 to perform the methods in the embodiments described above.
Furthermore, the logic instructions in the memory 22 may be implemented in software functional units and stored in a computer readable storage medium when sold or used as a stand-alone product.
The memory 22, which is a computer-readable storage medium, may be configured to store a software program, a computer-executable program, such as program instructions or modules corresponding to the methods in the embodiments of the present disclosure. The processor 20 executes the functional application and data processing, i.e. implements the method in the above-described embodiments, by executing the software program, instructions or modules stored in the memory 22.
The memory 22 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the electronic device, and the like. Further, the memory 22 may include a high speed random access memory and may also include a non-volatile memory. For example, a variety of media that can store program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, may also be transient storage media.
In addition, the specific processes loaded and executed by the storage medium and the instruction processors in the point cloud attribute encoding device are described in detail in the method, and are not stated herein.
In some embodiments, there is also provided a point cloud attribute decoding method, as shown in fig. 8, including the steps of:
s100, decoding the point cloud attribute code stream to obtain an attribute residual value of the current node;
s200, determining an attribute predicted value of the current node by adopting the point cloud attribute prediction method;
s300, determining the attribute value of the current node according to the sum of the attribute predicted value of the current node and the attribute residual value.
In this embodiment, a neighbor point which is physically closest to a current node is found by using an offset morton sequence, so as to determine an attribute predicted value of the current node, and finally, an attribute value of the current node is determined according to the attribute predicted value of the current node and a point cloud attribute code stream. The point cloud attribute decoding method provided by the invention can improve the utilization of the geometrical information and attribute information correlation of the point cloud, thereby improving the point cloud attribute decoding performance.
Based on the point cloud attribute decoding method, the invention also provides point cloud attribute decoding equipment, which comprises a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor; the communication bus realizes connection communication between the processor and the memory; the processor, when executing the computer readable program, implements the steps in the point cloud attribute decoding method of the present invention.
In some embodiments, the data obtained by comparing the result obtained by applying the point cloud attribute encoding method and the point cloud attribute decoding method provided by the present invention to the point cloud compression with the reference result of the test platform PCEM is shown in table 1-table 4, where the offset is (3, 3, 3), the search range is N-8, M-8:
TABLE 1 is a table comparing rate-distortion data for luminance, chrominance and reflectivity under lossy geometry and lossy properties
TABLE 2 is a table comparing rate-distortion data for luminance, chrominance and reflectivity under lossless geometry and lossy properties
TABLE 3 Rate distortion data comparison of luminance, chrominance and reflectivity under lossless geometry, finite loss property
TABLE 4 Rate distortion data comparison table for luminance, chrominance and reflectivity under lossless geometry, lossless property
As can be seen from the data in tables 1 to 4, compared with the reference result of the test platform PCEM, for the reflectivity attribute, under the conditions of the limited loss geometry and the lossy attribute, the conditions of the lossless geometry and the lossy attribute, and the conditions of the lossless geometry and the limited loss attribute, the end-to-end rate distortion of the invention is saved by 8.5%, 5.0% and 5.5% respectively; for brightness attribute, under the conditions of lossless geometry and lossless attribute, the code stream size is 94.7% of the original code stream size, under the conditions of limited loss geometry and lossy attribute, lossless geometry and lossy attribute, and the code stream size is 90% of the original code stream size, and the end-to-end rate distortion of the invention is respectively saved by 4.9%, 3.2% and 6.9%; for the chroma attribute, under the conditions of limited loss geometry, lossy attribute, lossless geometry and limited loss attribute, the end-to-end rate distortion of the invention is saved by 5.2%, 4.0% and 6.9% respectively at the highest.
In summary, the present invention determines the attribute prediction value of the current node by finding the nearest neighbor point in the physical space to the current node by using the offset morton sequence, and finally performs the attribute encoding according to the attribute prediction value or determines the attribute value of the current node according to the attribute prediction value of the current node and the point cloud attribute code stream. The point cloud attribute prediction method, the point cloud attribute encoding method and the point cloud attribute decoding method can improve the utilization of the geometrical information and attribute information correlation of the point cloud, so that the compression performance of the point cloud attributes is improved.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (9)
1. A point cloud attribute prediction method is characterized by comprising the following steps:
adding an offset value to the original point cloud coordinate to obtain a new coordinate value;
determining an offset morton sequence according to the new coordinate values;
and determining the attribute predicted value of the current node according to the offset morton sequence.
2. The method of point cloud attribute prediction according to claim 1, wherein the determining of the attribute prediction value of the current node according to the offset morton sequence comprises the steps of:
searching a first existing node forward according to the offset morton sequence to serve as a first prediction point;
taking the attribute value of the first prediction point as an attribute prediction value of the current node;
alternatively, K1 existing nodes are looked forward as first predicted points according to the offset morton sequence,
finding the point with the minimum distance from the current node at the K1 first predicted points;
taking the attribute value of the point with the minimum distance as an attribute predicted value of the current node;
alternatively, K1 existing nodes are looked forward as first predicted points according to the offset morton sequence,
and weighting the attribute values of the K1 first prediction points to be used as the attribute prediction value of the current node.
3. The method of point cloud attribute prediction according to claim 1, wherein the determining of the attribute prediction value of the current node according to the offset morton sequence comprises the steps of:
searching a first existing node forward according to the offset morton sequence to serve as a first prediction point;
searching a first existing node forward according to the original Morton sequence under the point cloud original coordinate to serve as a second prediction point;
calculating the distance d1 from the first predicted point to the current node;
calculating the distance d2 from the second predicted point to the current node;
comparing the distance d1 with the distance d2, and selecting a point with a small distance as a third predicted point;
taking the attribute value of the third prediction point as the attribute prediction value of the current node;
or, searching K1 existing nodes forward according to the offset Morton sequence as first prediction points;
searching K2 existing nodes forward according to the original Morton sequence under the point cloud original coordinate to serve as second prediction points;
calculating the distances from the first predicted point and the second predicted point to the current node, and selecting one or more points with small distances as third predicted points;
taking the attribute value or the attribute weighted value of the third predicted point as the attribute predicted value of the current node;
or, searching K1 existing nodes forward according to the offset Morton sequence as first prediction points;
searching K2 existing nodes forward according to the original Morton sequence under the point cloud original coordinate to serve as second prediction points;
calculating a weighted distance d1 from the first predicted point to the current node;
calculating a weighted distance d2 from the second predicted point to the current node;
comparing the distance d1 with the distance d2, and selecting a point with a small distance as a third predicted point;
and taking the attribute weighted value of the third prediction point as the attribute prediction value of the current node.
4. The point cloud attribute prediction method of claim 3, wherein the step of calculating the distance of the first predicted point to the current node comprises:
calculating the Euclidean distance between the current node and the first prediction point as the distance between the current node and the first prediction point;
or calculating the maximum value of the absolute value of the coordinate difference between the current node and the first prediction point in the X, Y and Z directions as the distance between the current node and the first prediction point;
or, calculating the sum of the difference values of the current node and the first prediction point coordinate in the X, Y and Z directions as the distance between the current node and the first prediction point.
5. The point cloud attribute prediction equipment is characterized by comprising a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor;
the communication bus realizes connection communication between the processor and the memory;
the processor, when executing the computer readable program, implements the steps in the point cloud attribute prediction method of any of claims 1-4.
6. A point cloud attribute coding method is characterized by comprising the following steps:
determining an attribute predicted value of a current node by adopting the point cloud attribute prediction method of any one of claims 1 to 4;
determining a prediction residual value of the current node according to a difference value between the attribute value of the current node and the attribute predicted value of the current node;
and coding the prediction residual value of the current node to obtain a point cloud code stream.
7. A point cloud attribute coding device is characterized by comprising a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor;
the communication bus realizes connection communication between the processor and the memory;
the processor, when executing the computer readable program, implements the steps in the point cloud attribute encoding method of claim 6.
8. A point cloud attribute decoding method is characterized by comprising the following steps:
decoding the point cloud attribute code stream to obtain an attribute residual value of the current node;
determining an attribute predicted value of a current node by adopting the point cloud attribute prediction method of any one of claims 1 to 4;
and determining the attribute value of the current node according to the sum of the attribute predicted value of the current node and the attribute residual value.
9. The point cloud attribute decoding equipment is characterized by comprising a processor, a memory and a communication bus; the memory has stored thereon a computer readable program executable by the processor;
the communication bus realizes connection communication between the processor and the memory;
the processor, when executing the computer readable program, implements the steps in the point cloud attribute decoding method of claim 8.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010239384.8A CN113473153B (en) | 2020-03-30 | 2020-03-30 | Point cloud attribute prediction method, encoding method, decoding method and equipment thereof |
| PCT/CN2021/083396 WO2021197238A1 (en) | 2020-03-30 | 2021-03-26 | Point cloud attribute prediction method and device, coding method and device, and decoding method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010239384.8A CN113473153B (en) | 2020-03-30 | 2020-03-30 | Point cloud attribute prediction method, encoding method, decoding method and equipment thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113473153A true CN113473153A (en) | 2021-10-01 |
| CN113473153B CN113473153B (en) | 2023-04-25 |
Family
ID=77865096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010239384.8A Active CN113473153B (en) | 2020-03-30 | 2020-03-30 | Point cloud attribute prediction method, encoding method, decoding method and equipment thereof |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113473153B (en) |
| WO (1) | WO2021197238A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024011739A1 (en) * | 2022-07-13 | 2024-01-18 | Oppo广东移动通信有限公司 | Point cloud encoding method, point cloud decoding method, codec and computer storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106126919A (en) * | 2016-06-23 | 2016-11-16 | 武汉大学 | A kind of accurate Hausdorff distance computational methods between any type point set data |
| CN107111489A (en) * | 2014-11-14 | 2017-08-29 | 英特尔公司 | Morton Coordinate Adjusting processor, method, system and instruction |
| CN107111485A (en) * | 2014-11-14 | 2017-08-29 | 英特尔公司 | Three-dimensional Morton Coordinate Conversion processor, method, system and instruction |
| US20190116372A1 (en) * | 2017-10-16 | 2019-04-18 | Mitsubishi Electric Research Laboratories, Inc. | Systems and Methods for Compressing, Representing and Processing Point Clouds |
| WO2019076503A1 (en) * | 2017-10-17 | 2019-04-25 | Nokia Technologies Oy | An apparatus, a method and a computer program for coding volumetric video |
| US20190228050A1 (en) * | 2018-01-19 | 2019-07-25 | 8i Limited | Bezier volume representation of point cloud attributes |
| CN110572655A (en) * | 2019-09-30 | 2019-12-13 | 北京大学深圳研究生院 | method and equipment for encoding and decoding point cloud attribute based on neighbor weight parameter selection and transmission |
| CN110580250A (en) * | 2018-06-08 | 2019-12-17 | 虚拟现实软件 | Indexing of mass point cloud data for efficient visualization |
| US20200021856A1 (en) * | 2018-07-10 | 2020-01-16 | Apple Inc. | Hierarchical point cloud compression |
| CN110708560A (en) * | 2018-07-10 | 2020-01-17 | 腾讯美国有限责任公司 | Point cloud data processing method and device |
| US20200043199A1 (en) * | 2018-07-13 | 2020-02-06 | Electronics And Telecommunications Research Institute | 3d point cloud data encoding/decoding method and apparatus |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110418135B (en) * | 2019-08-05 | 2022-05-27 | 北京大学深圳研究生院 | Point cloud intra-frame prediction method and device based on neighbor weight optimization |
| CN112218079B (en) * | 2020-08-24 | 2022-10-25 | 北京大学深圳研究生院 | Point cloud layering method based on spatial sequence, point cloud prediction method and point cloud prediction equipment |
-
2020
- 2020-03-30 CN CN202010239384.8A patent/CN113473153B/en active Active
-
2021
- 2021-03-26 WO PCT/CN2021/083396 patent/WO2021197238A1/en active Application Filing
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107111489A (en) * | 2014-11-14 | 2017-08-29 | 英特尔公司 | Morton Coordinate Adjusting processor, method, system and instruction |
| CN107111485A (en) * | 2014-11-14 | 2017-08-29 | 英特尔公司 | Three-dimensional Morton Coordinate Conversion processor, method, system and instruction |
| CN106126919A (en) * | 2016-06-23 | 2016-11-16 | 武汉大学 | A kind of accurate Hausdorff distance computational methods between any type point set data |
| US20190116372A1 (en) * | 2017-10-16 | 2019-04-18 | Mitsubishi Electric Research Laboratories, Inc. | Systems and Methods for Compressing, Representing and Processing Point Clouds |
| WO2019076503A1 (en) * | 2017-10-17 | 2019-04-25 | Nokia Technologies Oy | An apparatus, a method and a computer program for coding volumetric video |
| US20190228050A1 (en) * | 2018-01-19 | 2019-07-25 | 8i Limited | Bezier volume representation of point cloud attributes |
| CN110580250A (en) * | 2018-06-08 | 2019-12-17 | 虚拟现实软件 | Indexing of mass point cloud data for efficient visualization |
| US20200021856A1 (en) * | 2018-07-10 | 2020-01-16 | Apple Inc. | Hierarchical point cloud compression |
| CN110708560A (en) * | 2018-07-10 | 2020-01-17 | 腾讯美国有限责任公司 | Point cloud data processing method and device |
| US20200043199A1 (en) * | 2018-07-13 | 2020-02-06 | Electronics And Telecommunications Research Institute | 3d point cloud data encoding/decoding method and apparatus |
| CN110572655A (en) * | 2019-09-30 | 2019-12-13 | 北京大学深圳研究生院 | method and equipment for encoding and decoding point cloud attribute based on neighbor weight parameter selection and transmission |
Non-Patent Citations (1)
| Title |
|---|
| MICHAEL CONNOR 等: "Fast construction of k-nearest neighbor graphs for point clouds" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024011739A1 (en) * | 2022-07-13 | 2024-01-18 | Oppo广东移动通信有限公司 | Point cloud encoding method, point cloud decoding method, codec and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113473153B (en) | 2023-04-25 |
| WO2021197238A1 (en) | 2021-10-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111145090B (en) | Point cloud attribute coding method, point cloud attribute decoding method, point cloud attribute coding equipment and point cloud attribute decoding equipment | |
| CN111405281A (en) | Point cloud attribute information encoding method, point cloud attribute information decoding method, storage medium and terminal equipment | |
| US11699249B2 (en) | Trimming search space for nearest neighbor determinations in point cloud compression | |
| CN113256746B (en) | Method, device and storage medium for point cloud encoding and decoding | |
| CN112218079B (en) | Point cloud layering method based on spatial sequence, point cloud prediction method and point cloud prediction equipment | |
| CN111405284B (en) | Point cloud density-based attribute prediction method and device | |
| CN111242997A (en) | Filter-based point cloud attribute prediction method and device | |
| RU2505856C2 (en) | Method and apparatus for representing and identifying feature descriptors using compressed histogram of gradients | |
| US20200280718A1 (en) | Image encoding and decoding method and device | |
| KR20200039757A (en) | Point cloud compression | |
| CN109257604A (en) | A kind of color attribute coding method based on TMC3 point cloud encoder | |
| CN113473127B (en) | Point cloud geometric coding method, point cloud geometric decoding method, point cloud geometric coding equipment and point cloud geometric decoding equipment | |
| JP5911578B2 (en) | Method for encoding feature point position information of image, computer program, and mobile device | |
| CN113014903B (en) | Point cloud neighbor determination, point cloud prediction, point cloud coding and point cloud decoding methods and equipment | |
| CN104169946B (en) | Extensible queries for visual search | |
| CN112995662B (en) | Method and device for attribute entropy coding and entropy decoding of point cloud | |
| US20210104072A1 (en) | Techniques and apparatus for inter-channel prediction and transform for point-cloud attribute coding | |
| JP7430792B2 (en) | Attribute information prediction method, encoder, decoder and storage medium | |
| CN113473153B (en) | Point cloud attribute prediction method, encoding method, decoding method and equipment thereof | |
| CN104333758B (en) | The method and relevant apparatus of prediction technique and the detection pixel point of depth map | |
| CN113395506A (en) | Point cloud neighbor searching method, encoding method, decoding method and equipment based on segmentation | |
| KR20240006667A (en) | Point cloud attribute information encoding method, decoding method, device and related devices | |
| KR20230074504A (en) | Point cloud encoding method, point cloud decoding method and related device | |
| CN115412715B (en) | Method and device for predicting coding and decoding of point cloud attribute information | |
| Carr et al. | Solving multilabel graph cut problems with multilabel swap |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |