WO2020186548A1 - 点云编码和解码的方法、编码设备和解码设备 - Google Patents

点云编码和解码的方法、编码设备和解码设备 Download PDF

Info

Publication number
WO2020186548A1
WO2020186548A1 PCT/CN2019/079980 CN2019079980W WO2020186548A1 WO 2020186548 A1 WO2020186548 A1 WO 2020186548A1 CN 2019079980 W CN2019079980 W CN 2019079980W WO 2020186548 A1 WO2020186548 A1 WO 2020186548A1
Authority
WO
WIPO (PCT)
Prior art keywords
point
current point
value
distance
current
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.)
Ceased
Application number
PCT/CN2019/079980
Other languages
English (en)
French (fr)
Inventor
李革
魏红莲
邵薏婷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Peking University Shenzhen Graduate School
Original Assignee
Peking University Shenzhen Graduate School
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Peking University Shenzhen Graduate School filed Critical Peking University Shenzhen Graduate School
Priority to US17/279,177 priority Critical patent/US20220005230A1/en
Publication of WO2020186548A1 publication Critical patent/WO2020186548A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T9/00—Image coding
    • G06T9/001—Model-based coding, e.g. wire frame
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T9/00—Image coding
    • G06T9/004—Predictors, e.g. intraframe, interframe coding
    • 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
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/146—Data rate or code amount at the encoder output
    • 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
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • 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

Definitions

  • the present invention relates to the field of point cloud processing, and in particular to a method, encoding device and decoding device for point cloud encoding and decoding.
  • a point cloud is a set of discrete points that are randomly distributed in space and express the spatial structure and surface properties of a three-dimensional object or scene.
  • Each point in the point cloud has at least three-dimensional location information. Depending on the application scenario, it may also have color, material, or other information. Generally, each point in the point cloud has the same number of additional attributes.
  • Three-dimensional point cloud is an important form of digital representation of the real world. With the rapid development of 3D scanning equipment (laser, radar, etc.), the accuracy and resolution of point clouds are higher. High-precision point clouds are widely used in the construction of urban digital maps, and play a technical support role in many popular researches such as smart cities, unmanned driving, and cultural relics protection.
  • the point cloud is obtained by sampling the surface of an object by a three-dimensional scanning device.
  • the number of points in a frame of point cloud is generally in the order of one million. Each point contains geometric information, color, reflectivity and other attribute information, and the amount of data is very large.
  • the huge data volume of 3D point clouds brings huge challenges to data storage and transmission, so point cloud compression is very necessary.
  • Point cloud compression is mainly divided into geometric compression and attribute compression.
  • the test platform TMC13v5 (Test Model for Category 1&3 version 5) provided by the international standards organization MPEG (Moving Picture Experts Group) describes the point cloud attribute compression framework mainly:
  • Lifting strategy based on LOD Level of Detail: This method first constructs LOD with point clouds sorted according to Morton code, that is, set the distance threshold for each layer of LOD in advance, traverse all points in the point cloud, and meet The points of the threshold condition are added to the current LOD, and then the same method is used to determine the points that should be added to the next layer of LOD, until the entire LOD is constructed. Then find the 3 closest points for the points in the point cloud in LOD order according to the KNN algorithm, and use the weighted average of the attribute values of all 3 closest points as the predicted value.
  • LOD-based intra prediction mode decision-making method The process of constructing LOD in this method is the same as described in the above. After constructing LOD, according to KNN algorithm to find up to 3 neighbors for the current point, there are 3+1 prediction modes, respectively Yes: Use the weighted average of the attribute values of the 3 neighbors as the candidate predicted value, and use the attribute values of the first, second, and third neighbors as the candidate predicted value. Then calculate the RDO cost value for these 3+1 predicted values and corresponding modes, and use the corresponding predicted value of the smallest cost value as the predicted value of the current point.
  • the above related technologies only consider the distance and distance to predict the attribute value when determining the neighbor points, and the determined neighbor points are distributed on the same side of the current point, resulting in low prediction accuracy and reducing codec performance.
  • the present invention provides a point cloud coding and decoding method, coding equipment and decoding equipment, which can improve the point cloud coding and decoding performance by selecting appropriately distributed neighbor point sets.
  • a point cloud coding method which includes: determining a point set composed of K nearest neighbor points of the current point; determining a point set composed of L sub-nearest neighbor points of the current point; according to the K of the current point A point set consisting of two nearest neighbor points and a point set consisting of L next-neighbor points of the current point determine the preferred neighbor point set of the current point; encoding processing is performed according to the preferred neighbor point set of the current point.
  • the determining the point set composed of K nearest neighbors of the current point includes: according to the spatial distance between the point in the point cloud and the current point Determine the point set consisting of the K nearest neighbors of the current point.
  • the determining a point set consisting of L next-neighbor points of the current point includes: according to the The spatial distance from a point in the point cloud to the current point determines a point set consisting of L secondary neighbor points of the current point.
  • the point set composed of K nearest neighbors of the current point And the point set consisting of L next-neighbor points of the current point, determining the preferred neighbor point set of the current point includes: determining the distance from the centroid of the point set consisting of the K nearest neighbor points of the current point to the current point as The first distance; a point set consisting of K nearest neighbors of the current point, KH nearest neighbors and a point consisting of L second nearest neighbors of the current point, I sub-nearest points totaling K-H+I points The distance from the center of mass of the point set to the current point is determined as the second distance; when the difference between the first distance and the second distance is greater than the set value, a point consisting of K nearest neighbors of the current point A point set consisting of KH nearest neighbor points and L next neighbor points of the current point is determined as the preferred neighbor point set of the current point, or, When the difference between the first distance and
  • the point set composed of K nearest neighbors of the current point And the point set consisting of L next-neighbor points of the current point, determining the preferred neighbor point set of the current point includes: determining the distance from the centroid of the point set consisting of the K nearest neighbor points of the current point to the current point as The first distance; a point set consisting of K nearest neighbors of the current point, KH nearest neighbors and a point consisting of L second nearest neighbors of the current point, I sub-nearest points totaling K-H+I points The distance from the centroid of the point set to the current point is determined as the second distance; when the difference between the first distance and the second distance is less than or equal to the set value, the K nearest neighbors of the current point The composed point set is determined as the preferred neighbor point set of the current point, or, when the difference between the first distance and the second distance is greater than the set value: a point composed of K nearest neighbor points
  • the encoding process according to the preferred neighbor point set of the current point includes: the preferred neighbor point of the current point
  • the weighted average of the reconstructed attribute values of each point in the point set is determined as the candidate prediction value 1, and the corresponding mode is determined as the prediction mode 1.
  • the reconstructed attribute value of each point in the preferred neighbor point set of the current point is determined as the candidate prediction in turn From value 2 to candidate prediction value N+1, determine the corresponding modes in sequence from prediction mode 2 to prediction mode N+1, where N is the number of the point concentration; calculate the prediction mode 1 to the prediction mode N+1
  • the cost value of Jn, n is from 1 to N+1, the minimum value in Jn is determined as the minimum cost value, and the corresponding mode of the minimum cost value is determined as the optimal prediction mode;
  • the current point is determined according to the optimal prediction mode
  • the prediction value of the current point; the prediction residual value of the current point is determined according to the difference between the attribute value of the current point and the prediction value of the current point; the optimal prediction mode and the prediction residual value are encoded.
  • the encoding process according to the preferred neighbor point set of the current point includes: the preferred neighbor point of the current point
  • the weighted average of the reconstructed attribute values of each point in the point set is determined as the candidate prediction value 1, and the corresponding mode is determined as the prediction mode 1.
  • the reconstructed attribute value of each point in the preferred neighbor point set of the current point is determined as the candidate prediction in turn From value 2 to candidate prediction value N+1, determine the corresponding modes in sequence from prediction mode 2 to prediction mode N+1, where N is the number of the point concentration; calculate the prediction mode 1 to the prediction mode N+1
  • the cost value of Jn, n is from 1 to N+1, the minimum value in Jn is determined as the minimum cost value, and the corresponding mode of the minimum cost value is determined as the optimal prediction mode;
  • the current point is determined according to the optimal prediction mode
  • the predicted value of the current point; the predicted residual value of the current point is determined according to the difference between the attribute value of the current point and the predicted value of the current point; the difference between the first distance and the second distance is less than or equal to the setting
  • the optimal prediction mode and the prediction residual value are encoded, or the difference between the first distance and the second distance is greater than the set value and the second generation value is less than the first generation Value, encode the effective replacement flag, the optimal prediction mode and the
  • a point cloud decoding method including: determining a point set consisting of K nearest neighbors of a current point; determining a point set consisting of L next neighbors of the current point; A point set consisting of K nearest neighbor points and a point set consisting of L next-neighbor points of the current point determine a preferred neighbor point set of the current point; decoding processing is performed according to the preferred neighbor point set of the current point.
  • the determining the point set consisting of K nearest neighbors of the current point includes: according to the spatial distance between the point in the point cloud and the current point Determine the point set consisting of the K nearest neighbors of the current point.
  • the determining a point set consisting of L next-neighbor points of the current point includes: according to the spatial distance between the point in the point cloud and the current point The size determines the point set consisting of the L next neighbor points of the current point.
  • the point set composed of K nearest neighbors of the current point And the point set consisting of L next-neighbor points of the current point determining the preferred neighbor point set of the current point includes: determining the distance from the centroid of the point set consisting of the K nearest neighbor points of the current point to the current point as The first distance; a point set consisting of K nearest neighbors of the current point, KH nearest neighbors and a point consisting of L second nearest neighbors of the current point, I sub-nearest points totaling K-H+I points The distance between the centroid of the composed point set and the current point is determined as the second distance; when the difference between the first distance and the second distance is greater than the set value, the K nearest neighbor points of the current point are formed A point set consisting of KH nearest neighbor points and L next neighbor points of the current point in a point set consisting of I next neighbor points and a total of K-H+I points is determined
  • the point set composed of K nearest neighbors of the current point And the point set consisting of L next-neighbor points of the current point, determining the preferred neighbor point set of the current point includes: determining the distance from the centroid of the point set consisting of the K nearest neighbor points of the current point to the current point as The first distance; a point set consisting of K nearest neighbors of the current point, KH nearest neighbors and a point consisting of L second nearest neighbors of the current point, I sub-nearest points totaling K-H+I points The distance from the centroid of the point set to the current point is determined as the second distance; when the difference between the first distance and the second distance is less than or equal to the set value, the K nearest neighbors of the current point The composed point set is determined as the preferred neighbor point set of the current point, or, when the difference between the first distance and the second distance is greater than the set value and the replacement flag is a valid replacement flag
  • the decoding process according to the preferred neighbor point set of the current point includes: according to the optimal prediction mode and The preferred neighbor point set determines the predicted value of the current point; the attribute value of the current point is determined according to the sum of the predicted value of the current point and the predicted residual value of the current point.
  • the decoding processing according to the preferred neighbor point set of the current point includes: according to the optimal prediction mode and The preferred neighbor point set determines the predicted value of the current point; the attribute value of the current point is determined according to the sum of the predicted value of the current point and the predicted residual value of the current point.
  • a point cloud encoding device which includes: a first determining module for determining a point set consisting of K nearest neighbor points of a current point; a second determining module for determining L of the current point A point set consisting of several next-neighbor points; the third determining module is used to determine the current point’s current point based on the point set consisting of the current point’s K nearest neighbor points and the point set consisting of the current point’s L next-neighbor points The preferred neighboring point set; the encoding module is used to perform encoding processing according to the preferred neighboring point set of the current point.
  • the first determining module is specifically configured to: determine the K nearest points of the current point according to the spatial distance from the point in the point cloud to the current point A point set composed of neighboring points.
  • the second determining module is specifically configured to: determine L of the current point according to the spatial distance from the point in the point cloud to the current point The point set consisting of the next nearest neighbors.
  • the third determining module is specifically configured to: The distance from the centroid of the point set consisting of two nearest neighbors to the current point is determined as the first distance; the point set consisting of the K nearest neighbors of the current point has KH nearest neighbors and the L second nearest neighbors of the current point The distance between the center of mass of the point set consisting of I sub-neighbor points and K-H+I points in the point set of points to the current point is determined as the second distance; the difference between the first distance and the second distance is greater than When the setting value is set, the point consisting of K nearest neighbors of the current point is concentrated KH nearest neighbors and the point consisting of L secondary neighbors of the current point is concentrated I second nearest neighbors totaling K-H+I The point set consisting of points is determined as the preferred neighbor point set of the current point, or, when the difference between the first distance and the second distance is less than or equal to the set value, the K nearest neighbors of the
  • the third determining module is specifically configured to: The distance from the centroid of the point set consisting of two nearest neighbors to the current point is determined as the first distance; the point set consisting of the K nearest neighbors of the current point has KH nearest neighbors and the L second nearest neighbors of the current point The distance from the centroid of the point set consisting of I next-neighbor points and K-H+I points to the current point is determined as the second distance; the difference between the first distance and the second distance is less than Or equal to the set value, the point set consisting of the K nearest neighbor points of the current point is determined as the preferred neighbor point set of the current point, or the difference between the first distance and the second distance is greater than the When setting the value: determine the minimum cost value of the point set composed of K nearest neighbors of the current point as the first generation value; set KH nearest neighbors and the point set of K nearest neighbors of the current point The minimum cost value of the point set consisting of
  • the encoding module is specifically configured to: gather the current point’s preferred neighbor points to reconstruct the attribute value of each point The weighted average is determined as the candidate prediction value 1, and the corresponding mode is determined as the prediction mode 1. The reconstructed attribute value of each point in the preferred neighbor point set of the current point is determined as the candidate prediction value 2 to the candidate prediction value N+ 1.
  • the encoding module is specifically configured to: gather the current point’s preferred neighbor points to reconstruct the attribute values of each point The weighted average is determined as the candidate prediction value 1, and the corresponding mode is determined as the prediction mode 1. The reconstructed attribute value of each point in the preferred neighbor point set of the current point is determined as the candidate prediction value 2 to the candidate prediction value N+ 1.
  • a point cloud decoding device including: a first determining module for determining a point set composed of K nearest neighbors of a current point; a second determining module for determining L of the current point A point set composed of next-neighbor points; the third determining module is used to determine the preferred current point based on the point set composed of K nearest neighbor points of the current point and the point set composed of L second-neighbor points of the current point Neighbor point set; decoding module, used to decode the current point's preferred neighbor point set.
  • the first determining module is specifically configured to: determine the K nearest points of the current point according to the spatial distance from the point in the point cloud to the current point A point set composed of neighboring points.
  • the second determining module is specifically configured to: determine the L times of the current point according to the spatial distance from the point in the point cloud to the current point A point set consisting of neighboring points.
  • the third determining module is specifically configured to: The distance from the centroid of the point set consisting of two nearest neighbors to the current point is determined as the first distance; the point set consisting of the K nearest neighbors of the current point has KH nearest neighbors and the L second nearest neighbors of the current point The distance between the center of mass of the point set consisting of I sub-neighbor points and K-H+I points in the point set of points to the current point is determined as the second distance; the difference between the first distance and the second distance is greater than When the setting value is set, the point consisting of K nearest neighbors of the current point is concentrated KH nearest neighbors and the point consisting of L secondary neighbors of the current point is concentrated I second nearest neighbors totaling K-H+I The point set consisting of points is determined as the preferred neighbor point set of the current point, or, when the difference between the first distance and the second distance is less than or equal to the set value, the K nearest neighbors of the
  • the third determining module is specifically configured to: The distance from the centroid of the point set consisting of two nearest neighbors to the current point is determined as the first distance; the point set consisting of the K nearest neighbors of the current point has KH nearest neighbors and the L second nearest neighbors of the current point The distance from the centroid of the point set consisting of I next-neighbor points and K-H+I points to the current point is determined as the second distance; the difference between the first distance and the second distance is less than Or equal to the set value, the point set consisting of the K nearest neighbor points of the current point is determined as the preferred neighbor point set of the current point, or the difference between the first distance and the second distance is greater than the When the value is set and the replacement flag is the effective replacement flag, the point consisting of K nearest neighbors of the current point will be gathered KH nearest neighbors and the point consisting of L second neighbors of the current point will be gathered I second nearest neighbors.
  • the decoding module is specifically configured to: according to the optimal prediction mode and the preferred The set of neighboring points determines the predicted value of the current point; the attribute value of the current point is determined according to the sum of the predicted value of the current point and the predicted residual value of the current point.
  • the decoding module is specifically configured to: according to the optimal prediction mode and the preferred The set of neighboring points determines the predicted value of the current point; the attribute value of the current point is determined according to the sum of the predicted value of the current point and the predicted residual value of the current point.
  • the point cloud encoding and decoding method, encoding device, and decoding device of the embodiments of the present invention are based on a point set composed of K nearest neighbor points of the current point and a point set composed of L next neighbor points of the current point , Determine the preferred neighbor point set of the current point, thus, in the point cloud coding and decoding process, by selecting a properly distributed neighbor point set, the point cloud coding and decoding performance can be improved.
  • Fig. 1 is a schematic flowchart of a point cloud coding method according to an embodiment of the present invention
  • FIG. 2 is another schematic flowchart of a point cloud coding method according to an embodiment of the present invention.
  • Fig. 3 is a schematic diagram of a preferred neighbor point set according to an embodiment of the present invention.
  • Fig. 4 is another schematic flowchart of a point cloud coding method according to an embodiment of the present invention.
  • Fig. 5 is still another schematic flowchart of a point cloud coding method according to an embodiment of the present invention.
  • Fig. 6 is still another schematic flowchart of a point cloud coding method according to an embodiment of the present invention.
  • Fig. 7 is a schematic flowchart of a point cloud decoding method according to an embodiment of the present invention.
  • FIG. 8 is another schematic flowchart of a point cloud decoding method according to an embodiment of the present invention.
  • FIG. 9 is still another schematic flowchart of a point cloud decoding method according to an embodiment of the present invention.
  • Fig. 10 is still another schematic flowchart of a point cloud decoding method according to an embodiment of the present invention.
  • Figure 11 is a schematic block diagram of an encoding device according to an embodiment of the present invention.
  • Fig. 12 is a schematic block diagram of a decoding device according to an embodiment of the present invention.
  • Fig. 1 shows a schematic flowchart of a point cloud coding method according to an embodiment of the present invention.
  • the method shown in FIG. 1 may be executed by an encoding device, such as an encoder.
  • the method 1000 includes:
  • S1100 Determine a point set consisting of K nearest neighbor points of the current point
  • S1200 Determine a point set consisting of L next-neighbor points of the current point
  • S1300 Determine a preferred nearest neighbor point set of the current point according to a point set composed of K nearest neighbor points of the current point and a point set composed of L second nearest neighbor points of the current point;
  • S1400 Perform encoding processing according to the preferred neighbor point set of the current point.
  • the encoding device determines the point set consisting of K nearest neighbor points of the current point according to the spatial distance from the point in the point cloud to the current point, and determines the point consisting of L next neighbor points of the current point Set, according to the determined point set consisting of K nearest neighbor points of the current point and the point set consisting of L next neighbor points of the current point, determine the preferred neighbor point set of the current point, and proceed according to the determined preferred neighbor point set Encoding processing.
  • the encoding device determines the current point's current point by determining the point set consisting of K nearest neighbor points of the current point and the point set consisting of L next neighbor points of the current point.
  • the nearest neighbor point set is preferred, and encoding is performed according to the preferred neighbor point set. Therefore, during the encoding process, the preferred neighbor point set that is appropriately distributed can be selected to improve the coding performance.
  • the point set consisting of K nearest neighbor points of the current point is determined according to the spatial distance from the point in the point cloud to the current point.
  • the points in the point cloud are points that have been coded before the current point is coded.
  • the spatial distance may be Euclidean distance, cosine distance, or a distance calculated in other ways, which is not limited in the present invention.
  • the K is a natural number.
  • the point set composed of K nearest neighbor points of the current point is the point set composed of K points with the smallest distance between the point in the point cloud and the current point.
  • a point set consisting of L secondary neighbor points of the current point is determined according to the spatial distance from the point in the point cloud to the current point.
  • the points in the point cloud are points that have been coded before the current point is coded.
  • the spatial distance may be Euclidean distance, cosine distance, or a distance calculated in other ways, which is not limited in the present invention.
  • the L is a positive integer.
  • the L may be equal to the K or not equal to the K, which is not limited in the present invention.
  • the point set composed of L next-neighbor points of the current point is composed of L points in the point cloud with the smallest distance from the current point to the point after removing the K nearest neighbor points. Point set.
  • S1300 includes:
  • S1301 Determine the distance from the centroid of the point set consisting of K nearest neighbor points of the current point to the current point as the first distance;
  • the point consisting of the K nearest neighbors of the current point is collected by KH nearest neighbors and the point consisting of the L second nearest neighbors of the current point is collected by I second nearest neighbors, consisting of K-H+I points
  • the distance from the centroid of the point set to the current point is determined as the second distance
  • a point set consisting of K nearest neighbor points of the current point is determined as a preferred neighbor point set of the current point.
  • the H is a natural number and is less than or equal to the K.
  • the I is a positive integer and less than or equal to the L.
  • the I may be equal to the H or not equal to the H, which is not limited in the present invention.
  • the K-H nearest neighbor points are the K-H points with the smallest distance from the current point in the K nearest neighbor points of the current point.
  • the I second-neighbor points are the I points with the smallest distance from the current point in the point set consisting of the L second-neighbor points of the current point.
  • the set value may be a constant, such as 0, or other values, or may not be a constant, such as a function of the first distance and/or the second distance, such as multiplying the first distance With a coefficient, the present invention does not limit this.
  • FIG. 3 is a schematic diagram of a preferred neighbor point set according to an embodiment of the present invention.
  • the points P1, P2, and P3 in the three-dimensional space constitute the nearest neighbor point set of the current point O, and K is 3; the point P4 in the three-dimensional space constitutes the second nearest neighbor point set of the current point O, and L is 1;
  • point O1 is the point The centroid of the nearest neighbor point set composed of P1, P2 and P3;
  • point O2 is the centroid of the point set composed of points P1, P2 and P4, H is 1, I is 1;
  • the Euclidean distance between point O1 and point O is the same as point O2 When the difference of the Euclidean distance to point O is greater than 0, the point set consisting of points P1, P2, and P4 is determined as the preferred neighbor point set of the current point O.
  • the Euclidean distance from point O1 to point O is the same as the point O2 to point
  • the point set consisting of points P1, P2, and P3 is determined as the preferred neighbor point set of the current point O, and the set value is 0.
  • the coding device determines the preferred nearest neighbor point set of the current point by determining the point set consisting of K nearest neighbor points and the point set consisting of L second nearest neighbor points of the current point, The encoding process is performed according to the preferred neighboring point set of the current point, so that the encoding performance can be improved by selecting a suitably distributed preferred neighboring point set during the encoding process.
  • S1300 includes:
  • S1304 Determine the distance from the centroid of the point set consisting of K nearest neighbor points of the current point to the current point as the first distance;
  • the point set consisting of the K nearest neighbors of the current point is a point set consisting of KH nearest neighbors and a point set consisting of L second nearest neighbors of the current point.
  • a point set consisting of I second nearest neighbors and a total of K-H+I points The minimum generation value of is determined as the second generation value;
  • the point consisting of K nearest neighbors of the current point is concentrated to KH nearest neighbors and the point consisting of L second nearest neighbors of the current point is concentrated I times
  • the point set consisting of a total of K-H+I points of neighboring points is determined as the preferred neighboring point set of the current point, or,
  • the point set consisting of K nearest neighbor points of the current point is determined as the preferred neighbor point set of the current point.
  • the minimum cost value is further used to determine the preferred neighbor point set.
  • S1400 includes:
  • S1401 Determine the weighted average value of the reconstructed attribute values of each point in the preferred neighbor point set of the current point as candidate prediction value 1, and determine the corresponding mode as prediction mode 1.
  • S1402 Determine the reconstruction attribute value of each point in the preferred neighbor point set of the current point in sequence as candidate prediction value 2 to candidate prediction value N+1, and determine corresponding modes in sequence from prediction mode 2 to prediction mode N+1, Where N is the number of concentrated points at this point;
  • S1403 Calculate the cost value Jn from the prediction mode 1 to the prediction mode N+1, where n is from 1 to N+1, determine the minimum value in Jn as the minimum cost value, and determine the corresponding mode of the minimum cost value as Optimal forecasting mode;
  • S1405 Determine the predicted residual value of the current point according to the difference between the attribute value of the current point and the predicted value of the current point;
  • the reconstructed attribute value is a reconstructed attribute value after encoding.
  • the reconstructed attribute may be a color attribute, reflectance or other attributes, which is not limited in the present invention.
  • the weight in the weighted average value may be a function of the distance from each point in the point set to the current point as the weight.
  • the reciprocal of the distance may be the weight. It is other variables or constants, which are not limited by the present invention.
  • the method of encoding the optimal prediction mode and the prediction residual value may be arithmetic coding, Huffman coding or other entropy coding algorithms, and the present invention does not do this. limited.
  • Score is the total cost value
  • attrResidualQuant is the absolute value of the transformation and quantization of the difference between the attribute value of the current point and the predicted value in the prediction mode
  • idxBits*kAttrPredLambdaR*qs is the cost of encoding the prediction mode index, where idxBits is The prediction mode index, kAttrPredLambdaR is the preset constant, and qs is the attribute quantization step size.
  • the cost value calculation can also use other calculation methods, which are not limited in the present invention.
  • S1400 includes:
  • S1408 Determine the reconstructed attribute value of each point in the preferred neighbor point set of the current point in sequence as candidate prediction value 2 to candidate prediction value N+1, and determine corresponding modes in sequence from prediction mode 2 to prediction mode N+1, Where N is the number of concentrated points at this point;
  • S1409 Calculate the cost value Jn from the prediction mode 1 to the prediction mode N+1, where n is from 1 to N+1, determine the minimum value in Jn as the minimum cost value, and determine the corresponding mode of the minimum cost value as Optimal forecasting mode;
  • S1411 Determine the predicted residual value of the current point according to the difference between the attribute value of the current point and the predicted value of the current point;
  • the effective replacement flag, the optimal prediction mode, and the prediction residual value are encoded ,or
  • Table 1 shows the performance comparison result of the point cloud coding method in the embodiment of the present invention and the point cloud coding method in TMC13v5.
  • the encoding device determines the current point's current point by determining the point set consisting of K nearest neighbor points of the current point and the point set consisting of L next neighbor points of the current point.
  • the nearest neighbor point set is selected, and encoding is performed according to the preferred near neighbor point set. Therefore, during the encoding process, a suitably distributed preferred neighbor point set can be selected to improve coding performance.
  • Fig. 7 shows a schematic flowchart of a point cloud decoding method according to an embodiment of the present invention.
  • the method shown in FIG. 7 may be executed by a decoding device, such as a decoder.
  • the method 2000 includes:
  • S2100 Determine a point set consisting of K nearest neighbor points of the current point
  • S2200 Determine a point set composed of L next-neighbor points of the current point
  • S2300 Determine a preferred nearest neighbor point set of the current point according to a point set composed of K nearest neighbor points of the current point and a point set composed of L second nearest neighbor points of the current point;
  • S2400 Perform decoding processing according to the preferred neighbor point set of the current point.
  • the decoding device determines the point set composed of K nearest neighbor points of the current point according to the spatial distance from the point in the point cloud to the current point, and determines the point composed of L next neighbor points of the current point Set, according to the determined point set consisting of K nearest neighbor points of the current point and the point set consisting of L next neighbor points of the current point, determine the preferred neighbor point set of the current point, and proceed according to the determined preferred neighbor point set Decoding processing.
  • the decoding device determines the current point by determining the point set consisting of K nearest neighbor points of the current point and the point set consisting of L next neighbor points of the current point.
  • the nearest neighbor point set is preferred, and the decoding process is performed according to the preferred neighbor point set. Therefore, in the decoding process, a suitably distributed preferred neighbor point set can be selected to improve decoding efficiency.
  • a point set consisting of K nearest neighbor points of the current point is determined according to the spatial distance from a point in the decoded point cloud to the current point.
  • the spatial distance may be Euclidean distance or a distance calculated in other ways, which is not limited in the present invention.
  • the K is a natural number.
  • the point set composed of K nearest neighbor points of the current point is the point set composed of K points with the smallest distance between the point in the point cloud and the current point.
  • a point set consisting of L sub-neighbor points of the current point is determined according to the spatial distance from a point in the decoded point cloud to the current point.
  • the spatial distance may be Euclidean distance or a distance calculated in other ways, which is not limited in the present invention.
  • the L is a positive integer.
  • the L may be equal to the K or not equal to the K, which is not limited in the present invention.
  • the point set composed of L next-neighbor points of the current point is composed of L points in the point cloud with the smallest distance from the current point to the point after removing the K nearest neighbor points.
  • Point set is composed of L points in the point cloud with the smallest distance from the current point to the point after removing the K nearest neighbor points.
  • S2300 includes:
  • S2301 Determine the distance from the centroid of the point set consisting of K nearest neighbor points of the current point to the current point as the first distance;
  • a point set consisting of K nearest neighbors of the current point is composed of KH nearest neighbors and a point consisting of L second nearest neighbors of the current point, I sub-nearest points totaling K-H+I points The distance from the centroid of the point set to the current point is determined as the second distance;
  • a point set consisting of K nearest neighbor points of the current point is determined as a preferred neighbor point set of the current point.
  • the H is a natural number and is less than or equal to the K.
  • the I is a positive integer and less than or equal to the L.
  • the I may be equal to the H or not equal to the H, which is not limited in the present invention.
  • the K-H nearest neighbor points are the K-H points with the smallest distance from the current point in the K nearest neighbor points of the current point.
  • the I second-neighbor points are the I points with the smallest distance from the current point in the point set consisting of the L second-neighbor points of the current point.
  • the set value may be a constant, such as 0, or other values, or may not be a constant, such as a function of the first distance and/or the second distance, such as multiplying the first distance With a coefficient, the present invention does not limit this.
  • S2300 includes:
  • S2304 Determine the distance from the centroid of the point set consisting of K nearest neighbor points of the current point to the current point as the first distance;
  • a point set consisting of K nearest neighbors of the current point is composed of KH nearest neighbors and a point consisting of L second nearest neighbors of the current point.
  • I sub-nearest points are composed of K-H+I points. The distance from the centroid of the point set to the current point is determined as the second distance;
  • a point consisting of K nearest neighbors of the current point is concentrated on KH nearest neighbors and the current point
  • a point set consisting of a total of K-H+I points in a point set consisting of I secondary neighbor points and a total of K-H+I points is determined as the preferred neighbor point set of the current point, or,
  • the point set consisting of K nearest neighbor points of the current point is determined as the preferred neighbor point of the current point set.
  • the replacement flag is used to identify the effective replacement flag and the invalid replacement flag. It can be 1 bit of a binary bit, 1 represents a valid replacement flag, 0 represents an invalid replacement flag, or, conversely, 0 represents a valid replacement flag, and 1 represents The invalid replacement flag is not limited by the present invention.
  • S2400 includes:
  • S2401 Determine the prediction value of the current point according to the optimal prediction mode and the preferred neighbor point set;
  • S2402 Determine the attribute value of the current point according to the sum of the predicted value of the current point and the predicted residual value of the current point.
  • N+1 modes which are the weighted average mode of the reconstructed attribute values of each point in the point set, and the reconstructed attribute value mode of each point in the point set, where N is the value of the point in the point set.
  • the number is determined according to the optimal prediction mode, and then the prediction value of the current point is determined.
  • the encoding device 10 includes:
  • the first determining module 11 is configured to determine a point set composed of K nearest neighbor points of the current point;
  • the second determining module 12 is configured to determine a point set consisting of L next-neighbor points of the current point;
  • the third determining module 13 is configured to determine a preferred neighbor point set of the current point according to a point set composed of K nearest neighbor points of the current point and a point set composed of L secondary neighbor points of the current point;
  • the encoding module 14 is configured to perform encoding processing according to the preferred neighbor point set of the current point.
  • the encoding device determines the point set consisting of K nearest neighbor points of the current point according to the spatial distance from the point in the point cloud to the current point, and determines the point consisting of L next neighbor points of the current point Set, according to the determined point set consisting of K nearest neighbor points of the current point and the point set consisting of L next neighbor points of the current point, determine the preferred neighbor point set of the current point, and proceed according to the determined preferred neighbor point set Decoding processing.
  • the encoding device of the embodiment of the present invention determines the preferred neighbor point set of the current point by determining the point set consisting of K nearest neighbor points of the current point and the point set consisting of L second neighbor points of the current point, according to The preferred neighboring point set is encoded, so that during the encoding process, a suitably distributed preferred neighboring point set can be selected for encoding, thereby improving the encoding efficiency.
  • the decoding device 20 includes:
  • the first determining module 21 is configured to determine a point set composed of K nearest neighbor points of the current point;
  • the second determining module 22 is configured to determine a point set consisting of L next-neighbor points of the current point;
  • the third determining module 23 is configured to determine the preferred nearest neighbor point set of the current point according to the point set composed of K nearest neighbor points of the current point and the point set composed of L secondary neighbor points of the current point;
  • the decoding module 24 is configured to perform decoding processing according to the preferred neighbor point set of the current point.
  • the decoding device determines the point set composed of K nearest neighbor points of the current point according to the spatial distance from the point in the point cloud to the current point, and determines the point composed of L next neighbor points of the current point Set, according to the determined point set consisting of K nearest neighbor points of the current point and the point set consisting of L next neighbor points of the current point, determine the preferred neighbor point set of the current point, and proceed according to the determined preferred neighbor point set Decoding processing.
  • the decoding device in the embodiment of the present invention determines the preferred neighbor point set of the current point by determining the point set consisting of K nearest neighbor points of the current point and the point set consisting of L second neighbor points of the current point, according to The preferred neighboring point set is decoded, so that during the decoding process, a suitably distributed preferred neighboring point set can be selected for decoding, thereby improving decoding efficiency.
  • the size of the sequence numbers of the foregoing processes does not mean the order of execution.
  • the execution order of the processes should be determined by their functions and internal logic, and should not be used in the embodiments of the present invention.
  • the implementation process constitutes any limitation.
  • B corresponding to A means that B is associated with A, and B can be determined according to A.
  • determining B according to A does not mean that B is determined only according to A, and B can also be determined according to A and/or other information.
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • 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 system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional units in the various embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present invention essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention.
  • the three-dimensional point cloud of the present invention is an important manifestation of real world digitization. It is widely used in the construction of urban digital maps, and plays a technical support role in many popular researches such as smart cities, unmanned driving, and cultural relics protection.
  • the point cloud is obtained by sampling the surface of an object by a three-dimensional scanning device.
  • the number of points in a frame of point cloud is generally in the order of one million. Each point contains geometric information, color, reflectivity and other attribute information, and the amount of data is very large.
  • the huge data volume of 3D point clouds brings huge challenges to data storage and transmission, so point cloud compression is very necessary.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Abstract

本发明提供了一种点云编码和解码的方法、编码设备和解码设备。该方法包括：确定当前点的K个最近邻点组成的点集；确定该当前点的L个次近邻点组成的点集；根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集；根据所述当前点的优选近邻点集进行编码处理。由此，能够通过选取合适分布的近邻点集，提升点云编码性能。

Description

点云编码和解码的方法、编码设备和解码设备 技术领域
本发明涉及点云处理领域，尤其涉及点云编码和解码的方法、编码设备和解码设备。
背景技术
点云是空间中一组无规则分布的、表达三维物体或场景的空间结构及表面属性的离散点集。点云中的每个点至少具有三维位置信息，根据应用场景的不同，还可能具有色彩、材质或其他信息。通常，点云中的每个点都具有相同数量的附加属性。
三维点云是现实世界数字化的重要表现形式。随着三维扫描设备(激光、雷达等)的快速发展，点云的精度、分辨率更高。高精度点云广泛应用于城市数字化地图的构建，在如智慧城市、无人驾驶、文物保护等众多热门研究中起技术支撑作用。点云是三维扫描设备对物体表面采样所获取的，一帧点云的点数一般是百万级别，其中每个点包含几何信息和颜色、反射率等属性信息，数据量十分庞大。三维点云庞大的数据量给数据存储、传输等带来巨大挑战，所以点云压缩十分必要。
点云压缩主要分为几何压缩和属性压缩，目前由国际标准组织MPEG(Moving Picture Experts Group)所提供的测试平台TMC13v5(Test Model for Category 1&3 version 5)中描述的点云属性压缩框架主要有：
一、基于LOD(Level of Detail)的Lifting策略：该方法首先用已按照莫顿码排序的点云构建LOD，即对每一层LOD事先设置好距离阈值，遍历点云中的所有点，满足阈值条件的点加入当前LOD，然后再用相同方法判断应该加入下一层LOD的点，直至整个LOD构建完成。然后以LOD顺序对点云中的点按照KNN算法寻找3个最紧邻点，以所有3个最紧邻点的属性值的加权平均值作为预测值。
二、基于LOD的帧内预测模式决策方法：该方法构建LOD的过程同一中所述，构建完LOD之后，按照KNN算法为当前点寻找最多3个邻居，则共有3+1种预测模式，分别是：以3个邻居的属性值的加权平均值作为备选预测值，及以第1个、第2个、第3个邻居的属性值作为备选预测值。然后对此3+1个预测值及对应模式计算RDO代价值，将其中最小代价值的对应预测值作为当前点的预测值。
但是，以上相关技术在近邻点的确定时只考虑了距离远近对属性值进行预测，存在确定的近邻点分布在当前点的同侧，导致预测准确度不高，降低编解码性能。
发明的公开
本发明提供了一种点云编码和解码的方法、编码设备和解码设备，可以通过选取合适分布的近邻点集，提升点云编解码性能。
第一方面，提供了一种点云编码方法，包括：确定当前点的K个最近邻点组成的点集；确定该当前点的L个次近邻点组成的点集；根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集；根据该当前点的优选近邻点集进行编码处理。
结合第一方面，在第一方面的第一种可能的实现方式中，该确定当前点的K个最近邻点组成的点集，包括：根据点云中的点到该当前点的空间距离大小确定该当前点的K个最近邻点组成的点集。
结合第一方面，第一方面的第一种可能的实现方式，在第一方面的第二种可能的实现方式中，该确定当前点的L个次近邻点组成的点集，包括：根据该点云中的点到该当前点的空间距离大小确定该当前点的L个次近邻点组成的点集。
结合第一方面，第一方面的第一种至第二种可能的实现方式，在第一方面的第三种可能的实现方式中，该根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值大于设定值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第一方面，第一方面的第一种至第二种可能的实现方式，在第一方面的第四种可能的实现方式中，该根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值大于该设定值时：将该当前点的K个最近邻点组成的点集的最小代价值确定为第一代价值；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的最小代价值确定为第二代价值；在该 第二代价值小于该第一代价值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第二代价值大于或等于该第一代价值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第一方面的第三种可能的实现方式，在第一方面的第五种可能的实现方式中，该根据该当前点的优选近邻点集进行编码处理，包括：将该当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；依次将该当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为该点集中点的个数；计算该预测模式1到该预测模式N+1的代价值Jn，n由1到N+1，将该Jn中的最小值确定为最小代价值，将该最小代价值的对应模式确定为最优预测模式；根据该最优预测模式确定当前点的预测值；根据该当前点的属性值与该当前点的预测值之间的差值确定当前点的预测残差值；对该最优预测模式及所述预测残差值进行编码。
结合第一方面的第四种可能的实现方式，在第一方面的第六种可能的实现方式中，该根据该当前点的优选近邻点集进行编码处理，包括：将该当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；依次将该当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为该点集中点的个数；计算该预测模式1到该预测模式N+1的代价值Jn，n由1到N+1，将该Jn中的最小值确定为最小代价值，将该最小代价值的对应模式确定为最优预测模式；根据该最优预测模式确定当前点的预测值；根据该当前点的属性值与该当前点的预测值之间的差值确定当前点的预测残差值；在该第一距离与该第二距离的差值小于或等于该设定值时，对该最优预测模式及该预测残差值进行编码，或，在该第一距离与该第二距离的差值大于该设定值且该第二代价值小于该第一代价值时，对有效替换标志、该最优预测模式及该预测残差值进行编码，或，在该第一距离与该第二距离的差值大于该设定值且该第二代价值大于或等于该第一代价值时，对无效替换标志、该最优预测模式及该预测残差值进行编码。
第二方面，提供了一种点云解码方法，包括：确定当前点的K个最近邻点组成的点集；确定所述当前点的L个次近邻点组成的点集；根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集；根据所述当前点的优选近邻点集进行解码处理。
结合第二方面，在第二方面的第一种可能的实现方式中，该确定当前点的K个最近邻点组成的点集，包括：根据点云中的点到该当前点的空间距离大小确定该当前点的K个最近邻点组成的点集。
结合第二方面，在第二方面的第二种可能的实现方式中，该确定当前点的L个次近邻点组成的点集，包括：根据该点云中的点到该当前点的空间距离大小确定该当前点的L个次近邻点组成的点集。
结合第二方面，第二方面的第一种至第二种可能的实现方式，在第二方面的第三种可能的实现方式中，该根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值大于该设定值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第二方面，第二方面的第一种至第二种可能的实现方式，在第二方面的第四种可能的实现方式中，该根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值大于该设定值且替换标志为有效替换标志时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值大于该设定值且替换标志为无效替换标志时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第二方面的第三种可能的实现方式，在第二方面的第五种可能的实现方式中，该根据该当前点的优选近邻点集进行解码处理，包括：根据该最优预测模式及该优选近邻点集确定当前点的预测值；根据该当前点的预测值与该当前点的预测残差值的和确定该当前点的属性值。
结合第二方面的第四种可能的实现方式，在第二方面的第六种可能的实现方式中，该根据该当前点的优选近邻点集进行解码处理，包括：根据该最优预测模式及该优选近邻点集确定当前点的预测值；根据该当前点的预测值与该当前点的预测残差值的和确定该当前点的属性值。
第三方面，提供了一种点云编码设备，包括：第一确定模块，用于确定当前点的K个最近邻点组成的点集；第二确定模块，用于确定所述当前点的L个次近邻点组成的点集；第三确定模块，用于根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集；编码模块，用于根据该当前点的优选近邻点集进行编码处理。
结合第三方面，在第三方面的第一种可能的实现方式中，该第一确定模块具体用于：根据点云中的点到该当前点的空间距离大小确定该当前点的K个最近邻点组成的点集。
结合第三方面，在第三方面的第二种可能的实现方式中，该第二确定模块具体用于：根据该点云中的点到该当前点的空间距离大小确定该当前点的L个次近邻点组成的点集。
结合第三方面，第三方面的第一种至第二种可能的实现方式，在第三方面的第三种可能的实现方式中，该第三确定模块具体用于：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值大于该设定值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第三方面，第三方面的第一种至第二种可能的实现方式，在第三方面的第四种可能的实现方式中，该第三确定模块具体用于：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值大于该设定值时：将该当前点的K个最近邻点组成的点集的最小代价值确定为第一代价值；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的最小代价值确定为第二代价值；在该第二代价值小于该第一代价值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第二代价值大于或等于该第一代价值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第三方面的第三种可能的实现方式，在第三方面的第五种可能的实现方式中，该编码模块具体用于：将该当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；依次将该当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为该点集中点的个数；计算该预测模式1到该预测模式N+1的代价值Jn，n由1到N+1，将该Jn中的最小值确定为最小代价值，将该最小代价值的对应模式确定为最优预测模式；根据该最优预测模式确定当前点的预测值；根据该当前点的属性值与该当前点的预测值之间的差值确定当前点的预测残差值；对该最优预测模式及所述预测残差值进行编码。
结合第三方面的第四种可能的实现方式，在第三方面的第六种可能的实现方式中，该编码模块具体用于：将该当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；依次将该当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为该点集中点的个数；计算该预测模式1到该预测模式N+1的代价值Jn，n由1到N+1，将该Jn中的最小值确定为最小代价值，将该最小代价值的对应模式确定为最优预测模式；根据该最优预测模式确定当前点的预测值；根据该当前点的属性值与该当前点的预测值之间的差值确定当前点的预测残差值；在该第一距离与该第二距离的差值小于或等于该设定值时，对该最优预测模式及该预测残差值进行编码，或，在该第一距离与该第二距离的差值大于该设定值且该第二代价值小于该第一代价值时，对有效替换标志、该最优预测模式及该预测残差值进行编码，或，在该第一距离与该第二距离的差值大于该设定值且该第二代价值大于或等于该第一代价值时，对无效替换标志、该最优预测模式及该预测残差值进行编码。
第四方面，提供了一种点云解码设备，包括：第一确定模块，用于确定当前点的K个最近邻点组成的点集；第二确定模块，用于确定该当前点的L个次近邻点组成的点集；第三确定模块，用于根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集；解码模块，用于根据该当前点的优选近邻点集进行解码处理。
结合第四方面，在第四方面的第一种可能的实现方式中，该第一确定模块具体用于：根据点云中的点到该当前点的空间距离大小确定该当前点的K个最近邻点组成的点集。
结合第四方面在第四方面的第二种可能的实现方式中，该第二确定模块具体用于：根据该点云中的点到该当前点的空间距离大小确定该当前点的L个次近邻点组成的点集。
结合第四方面，第四方面的第一种至第二种可能的实现方式，在第四方面的第三 种可能的实现方式中，该第三确定模块具体用于：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值大于该设定值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第四方面，第四方面的第一种至第二种可能的实现方式，在第四方面的第四种可能的实现方式中，该第三确定模块具体用于：将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值大于该设定值且替换标志为有效替换标志时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，在该第一距离与该第二距离的差值大于该设定值且替换标志为无效替换标志时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
结合第四方面的第三种可能的实现方式中任一可能的实现方式，在第四方面的第五种可能的实现方式中，该解码模块具体用于：根据该最优预测模式及该优选近邻点集确定当前点的预测值；根据该当前点的预测值与该当前点的预测残差值的和确定该当前点的属性值。
结合第四方面的第四种可能的实现方式中任一可能的实现方式，在第四方面的第六种可能的实现方式中，该解码模块具体用于：根据该最优预测模式及该优选近邻点集确定当前点的预测值；根据该当前点的预测值与该当前点的预测残差值的和确定该当前点的属性值。
基于上述技术方案，本发明实施例的点云编码和解码的方法、编码设备和解码设备，根据当前点的K个最近邻点组成的点集及当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，由此，在点云编解码过程中能够通过选取合适分布的近邻点集，提升点云编解码性能。
附图的简要说明
为了更清楚地说明本发明实施例的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是根据本发明实施例的点云编码方法的示意性流程图；
图2是根据本发明实施例的点云编码方法的另一示意性流程图；
图3是根据本发明实施例的优选近邻点集的示意图；
图4是根据本发明实施例的点云编码方法的再一示意性流程图；
图5是根据本发明实施例的点云编码方法的再一示意性流程图；
图6是根据本发明实施例的点云编码方法的再一示意性流程图；。
图7是根据本发明实施例的点云解码方法的示意性流程图；
图8是根据本发明实施例的点云解码方法的另一示意性流程图；
图9是根据本发明实施例的点云解码方法的再一示意性流程图；
图10是根据本发明实施例的点云解码方法的再一示意性流程图；
图11是根据本发明实施例的编码设备的示意性框图；
图12是根据本发明实施例的解码设备的示意性框图。
实现本发明的最佳方式
下面结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有付出创造性劳动的前提下所获得的所有其他实施例，都属于本发明保护的范围。
图1示出了根据本发明实施例的点云编码方法的示意性流程图。图1所示的方法可以由编码设备执行，例如编码器，具体地，如图1所示，方法1000包括：
S1100，确定当前点的K个最近邻点组成的点集；
S1200，确定该当前点的L个次近邻点组成的点集；
S1300，根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集；
S1400，根据该当前点的优选近邻点集进行编码处理。
具体而言，编码设备根据点云中的点到所述当前点的空间距离大小，确定当前点的K个最近邻点组成的点集，并确定该当前点的L个次近邻点组成的点集，根据确定的当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据确定的优选近邻点集进行编码处理。
因此，本发明实施例的点云编码的方法，编码设备通过确定的当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据该优选近邻点集进行编码处理，由此，在编码过程中能够通过选取合适分布的该优选近邻点集，提高编码性能。
优选地，在S1100中，根据点云中的点到该当前点的空间距离大小确定该当前点的K个最近邻点组成的点集。
应理解，在本发明实施例中，该点云中的点为在当前点编码之前已编码过的点。
还应理解，在本发明实施例中，该空间距离可以是欧式距离、余弦距离或者其它方式计算的距离，本发明对此不作限定。
还应理解，在本发明实施例中，该K为自然数。
还应理解，在本发明实施例中，该当前点的K个最近邻点组成的点集为该点云中的点与当前点的距离最小的K个点组成的点集。
优选地，在S1200中，根据该点云中的点到该当前点的空间距离大小确定该当前点的L个次近邻点组成的点集。
应理解，在本发明实施例中，该点云中的点为在当前点编码之前已编码过的点。
还应理解，在本发明实施例中，该空间距离可以是欧式距离、余弦距离或者其它方式计算的距离，本发明对此不作限定。
还应理解，在本发明实施例中，该L为正整数。
还应理解，在本发明实施例中，该L可以等于该K，也可以不等于该K，本发明对此不作限定。
还应理解，在本发明实施例中，该当前点的L个次近邻点组成的点集为该点云中除去该K个最近邻点后的点与当前点的距离最小的L个点组成的点集。
优选地，如图2所示，S1300包括：
S1301，将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；
S1302，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及所该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前 点的距离确定为第二距离；
S1303，在该第一距离与该第二距离的差值大于该设定值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，
在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
应理解，在本发明实施例中，该H为自然数且小于或等于该K。
还应理解，在本发明实施例中，该I为正整数且小于或等于该L。
还应理解，在本发明实施例中，该I可以等于该H，也可以不等于该H，本发明对此不作限定。
还应理解，在本发明实施例中，该K-H个最近邻点为该当前点的K个最近邻点集中与当前点的距离最小的K-H个点。
还应理解，在本发明实施例中，该I个次近邻点为该当前点的L个次近邻点组成的点集中与当前点的距离最小的I个点。
还应理解，在本发明实施例中，该设定值可以是常数，如0，或其它数值，也可以不是常数，如第一距离和/或第二距离的一个函数，如第一距离乘以一个系数，本发明对此不作限定。
例如，图3是根据本发明实施例的优选近邻点集的示意图。其中，三维空间中的点P1、P2及P3组成当前点O的最近邻点集，K为3；三维空间中的点P4组成当前点O的次近邻点集，L为1；点O1为点P1、P2及P3组成的最近邻点集的质心；点O2为点P1、P2及P4组成的点集的质心，H为1，I为1；在点O1到点O的欧式距离与点O2到点O的欧式距离的差值大于0时，将点P1、P2及P4组成的点集确定为该当前点O的优选近邻点集，在点O1到点O的欧式距离与点O2到点O的欧式距离的差值小于或等于0时，将点P1、P2及P3组成的点集确定为该当前点O的优选近邻点集，其中设定值为0。
因此，本发明实施例的点云编码的方法，编码设备通过确定的当前点的K个最近邻点组成的点集及L个次近邻点组成的点集，确定当前点的优选近邻点集，根据该当前点的优选近邻点集进行编码处理，由此，在编码过程中能够通过选取合适分布的优选近邻点集，提高编码性能。
可选地，如图4所示，S1300包括：
S1304，将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；
S1305，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及所该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；
S1306，在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集，或，
在该第一距离与该第二距离的差值大于该设定值时：
将该当前点的K个最近邻点组成的点集的最小代价值确定为第一代价值；
将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的最小代价值确定为第二代价值；
在该第二代价值小于该第一代价值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，
在该第二代价值大于或等于该第一代价值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
具体而言，在该第一距离与该第二距离的差值大于该设定值时，进一步通过最小代价值确定优选近邻点集。
优选地，如图5所示，S1400包括：
S1401，将该当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；
S1402，依次将该当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为该点集中点的个数；
S1403，计算该预测模式1到该预测模式N+1的代价值Jn，n由1到N+1，将该Jn中的最小值确定为最小代价值，将该最小代价值的对应模式确定为最优预测模式；
S1404，根据该最优预测模式确定当前点的预测值；
S1405，根据该当前点的属性值与该当前点的预测值之间的差值确定当前点的预测残差值；
S1406，对该最优预测模式及所述预测残差值进行编码。
应理解，在本发明实施例中，该重建属性值是经过编码后的重建的属性值，该重建属性可以是颜色属性，也可以是反射率或者其它属性，本发明对此不作限定。
还应理解，在本发明实施例中，该加权平均值中的权值，可以是以点集中的各点到该当前点的距离的函数为权值，如距离的倒数为权值，也可以是其它变量或常量，本发明对此不作限定。
还应理解，在本发明实施例中，对该最优预测模式及所述预测残差值进行编码的方式可以是算数编码，也可以是哈弗曼编码或者其它熵编码算法，本发明对此不作限定。
还应理解，在本发明实施例中，代价值的一种计算方式为：
score＝attrResidualQuant+idxBits*kAttrPredLambdaR*qs
score为总代价值；attrResidualQuant为该预测模式下，当前点的属性值与预测值的差值经过变换、量化后的绝对值；idxBits*kAttrPredLambdaR*qs为编码该预测模式索引的代价，其中idxBits是该预测模式索引，kAttrPredLambdaR为预设常量，qs是属性量化步长。代价值的计算也可以采用其它计算方式，本发明对此不作限定。
可选地，如图6所示，S1400包括：
S1407，将该当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；
S1408，依次将该当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为该点集中点的个数；
S1409，计算该预测模式1到该预测模式N+1的代价值Jn，n由1到N+1，将该Jn中的最小值确定为最小代价值，将该最小代价值的对应模式确定为最优预测模式；
S1410，根据该最优预测模式确定当前点的预测值；
S1411，根据该当前点的属性值与该当前点的预测值之间的差值确定当前点的预测残差值；
S1412，在该第一距离与该第二距离的差值小于或等于该设定值时，对该最优预测模式及该预测残差值进行编码，或，
在该第一距离与该第二距离的差值大于该设定值且该第二代价值小于该第一代价值时，对有效替换标志、该最优预测模式及该预测残差值进行编码，或，
在该第一距离与该第二距离的差值大于该设定值且该第二代价值大于或等于该第一代价值时，对无效替换标志、该最优预测模式及该预测残差值进行编码。为了更好的证明本发明实施例相对于现有技术的有益效果，表1展示了本发明实施例的点云编码方法与TMC13v5中的点云编码方法的性能对比结果。
表1
Figure PCTCN2019079980-appb-000001
通过表1可以看写出，采用本发明实施例的点云编码的方法，能够提升编码性。
因此，本发明实施例的点云编码的方法，编码设备通过确定的当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据该优选近邻点集进行编码处理，由此，在编码过程中能够通过选取合适分布的优选近邻点集，提高编码性能。
以上结合图1至图6详细描述了根据本发明实施例的点云编码的方法，下面将结合图7至图10详细描述根据本发明实施例的点云解码的方法。应注意，解码端的有关操作本质上与编码端的一致，为了避免重复，在此不再赘述。
图7示出了根据本发明实施例的点云解码的方法的示意性流程图。图7所示的方法可以由解码设备执行，例如解码器，具体地，如图7所示，方法2000包括：
S2100，确定当前点的K个最近邻点组成的点集；
S2200，确定所述当前点的L个次近邻点组成的点集；
S2300，根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集；
S2400，根据所述当前点的优选近邻点集进行解码处理。
具体而言，解码设备根据点云中的点到所述当前点的空间距离大小，确定当前点的K个最近邻点组成的点集，并确定该当前点的L个次近邻点组成的点集，根据确定的当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据确定的优选近邻点集进行解码处理。
因此，本发明实施例的点云解码的方法，解码设备通过确定的当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据该优选近邻点集进行解码处理，由此，在解码过程中能够通过选取合适分布 的优选近邻点集，提高解码效率。
优选地，在S2100中，根据已解码点云中的点到该当前点的空间距离大小确定该当前点的K个最近邻点组成的点集。
还应理解，在本发明实施例中，该空间距离可以是欧式距离，或者其它方式计算的距离，本发明对此不作限定。
还应理解，在本发明实施例中，该K为自然数。
还应理解，在本发明实施例中，该当前点的K个最近邻点组成的点集为该点云中的点与当前点的距离最小的K个点组成的点集。
优选地，在S2200中，根据已解码点云中的点到该当前点的空间距离大小确定该当前点的L个次近邻点组成的点集。
还应理解，在本发明实施例中，该空间距离可以是欧式距离，或者其它方式计算的距离，本发明对此不作限定。
还应理解，在本发明实施例中，该L为正整数。
还应理解，在本发明实施例中，该L可以等于该K，也可以不等于该K，本发明对此不作限定。
还应理解，在本发明实施例中，该当前点的L个次近邻点组成的点集为该点云中除去该K个最近邻点后的点与当前点的距离最小的L个点组成的点集。优选地，如图8所示，S2300包括：
S2301，将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；
S2302，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；
S2303，在该第一距离与该第二距离的差值大于该设定值时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，
在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
应理解，在本发明实施例中，该H为自然数且小于或等于该K。
还应理解，在本发明实施例中，该I为正整数且小于或等于该L。
还应理解，在本发明实施例中，该I可以等于该H，也可以不等于该H，本发明 对此不作限定。
还应理解，在本发明实施例中，该K-H个最近邻点为该当前点的K个最近邻点集中与当前点的距离最小的K-H个点。
还应理解，在本发明实施例中，该I个次近邻点为该当前点的L个次近邻点组成的点集中与当前点的距离最小的I个点。
还应理解，在本发明实施例中，该设定值可以是常数，如0，或其它数值，也可以不是常数，如第一距离和/或第二距离的一个函数，如第一距离乘以一个系数，本发明对此不作限定。
可选的，如图9所示，S2300包括：
S2304，将该当前点的K个最近邻点组成的点集的质心到该当前点的距离确定为第一距离；
S2305，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到该当前点的距离确定为第二距离；
S2306，在该第一距离与该第二距离的差值小于或等于该设定值时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集，或，
在该第一距离与该第二距离的差值大于该设定值且替换标志为有效替换标志时，将该当前点的K个最近邻点组成的点集中K-H个最近邻点及该当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为该当前点的优选近邻点集，或，
在该第一距离与该第二距离的差值大于该设定值且替换标志为无效替换标志时，将该当前点的K个最近邻点组成的点集确定为该当前点的优选近邻点集。
应理解，替换标志是用来标识有效替换标志与无效替换标志的，可以是二进制位的1位，1表示有效替换标志，0表示无效替换标志，或者，反过来0表示有效替换标志，1表示无效替换标志，，本发明对此不作限定。
优选地，如图10所示，S2400包括：
S2401，根据该最优预测模式及该优选近邻点集确定当前点的预测值；
S2402，根据该当前点的预测值与该当前点的预测残差值的和确定该当前点的属性值。
优选地，对应于一个点集，存在N+1种模式，分别是点集中各点的重建属性值的加权平均值模式、点集中各点的重建属性值模式，其中N为该点集中点的个数，根据 该最优预测模式确定是其中的哪一种模式，进而确定当前点的预测值。
下面将结合图11详细描述本发明实施例的编码设备。如图11所示，该编码设备10包括：
第一确定模块11，用于确定当前点的K个最近邻点组成的点集；
第二确定模块12，用于确定所述当前点的L个次近邻点组成的点集；
第三确定模块13，用于根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集；
编码模块14，用于根据该当前点的优选近邻点集进行编码处理。
具体而言，编码设备根据点云中的点到所述当前点的空间距离大小，确定当前点的K个最近邻点组成的点集，并确定该当前点的L个次近邻点组成的点集，根据确定的当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据确定的优选近邻点集进行解码处理。
因此，本发明实施例的编码设备通过确定的当前点的K个最近邻点组成的点集和该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据该优选近邻点集进行编码处理，由此，在编码过程中能够通过选取合适分布的优选近邻点集进行编码，提高编码效率。
下面将结合图12详细描述本发明实施例的解码设备。如图12所示，该解码设备20包括：
第一确定模块21，用于确定当前点的K个最近邻点组成的点集；
第二确定模块22，用于确定该当前点的L个次近邻点组成的点集；
第三确定模块23，用于根据该当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集；
解码模块24，用于根据该当前点的优选近邻点集进行解码处理。
具体而言，解码设备根据点云中的点到所述当前点的空间距离大小，确定当前点的K个最近邻点组成的点集，并确定该当前点的L个次近邻点组成的点集，根据确定的当前点的K个最近邻点组成的点集及该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据确定的优选近邻点集进行解码处理。
因此，本发明实施例的解码设备通过确定的当前点的K个最近邻点组成的点集和该当前点的L个次近邻点组成的点集，确定该当前点的优选近邻点集，根据该优选近邻点集进行解码处理，由此，在解码过程中能够通过选取合适分布的优选近邻点集进行解码，提高解码效率。
应理解，在本发明的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本发明实施例的实施过程构成任何限定。
应理解，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
在本申请所提供的实施例中，应理解，“与A相应的B”表示B与A相关联，根据A可以确定B。但还应理解，根据A确定B并不意味着仅仅根据A确定B，还可以根据A和/或其它信息确定B。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出 来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以所述权利要求的保护范围为准。
工业应用性
本发明的三维点云是现实世界数字化的重要表现形式。广泛应用于城市数字化地图的构建，在如智慧城市、无人驾驶、文物保护等众多热门研究中起技术支撑作用。点云是三维扫描设备对物体表面采样所获取的，一帧点云的点数一般是百万级别，其中每个点包含几何信息和颜色、反射率等属性信息，数据量十分庞大。三维点云庞大的数据量给数据存储、传输等带来巨大挑战，所以点云压缩十分必要。

Claims (26)

  1. 一种点云编码方法，其特征在于，包括：
    确定当前点的K个最近邻点组成的点集；
    确定所述当前点的L个次近邻点组成的点集；
    根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集；
    根据所述当前点的优选近邻点集进行编码处理。
  2. 根据权利要求1所述的方法，其特征在于，所述确定当前点的K个最近邻点组成的点集，包括：
    根据点云中的点到所述当前点的空间距离大小确定所述当前点的K个最近邻点组成的点集。
  3. 根据权利要求1所述的方法，其特征在于，所述确定当前点的L个次近邻点组成的点集，包括：
    根据所述点云中的点到所述当前点的空间距离大小确定所述当前点的L个次近邻点组成的点集。
  4. 根据权利要求1至3所述的方法，其特征在于，所述根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值大于设定值时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集；或，
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  5. 根据权利要求1至3所述的方法，其特征在于，所述根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值大于所述设定值时：
    将所述当前点的K个最近邻点组成的点集的最小代价值确定为第一代价值；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的最小代价值确定为第二代价值；
    在所述第二代价值小于所述第一代价值时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第二代价值大于或等于所述第一代价值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  6. 根据权利要求4所述的方法，其特征在于，所述根据所述当前点的优选近邻点集进行编码处理，包括：
    将所述当前点的优选近邻点集中各点的重建属性值的加权平均值确定为预测值1，将对应模式确定为预测模式1；
    依次将所述当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为所述点集中点的个数；
    计算所述预测模式1到所述预测模式N+1的代价值Jn，n由1到N+1，将所述Jn中的最小值确定为最小代价值，将所述最小代价值的对应模式确定为最优预测模式；
    根据所述最优预测模式确定当前点的预测值；
    根据所述当前点的属性值与所述当前点的预测值之间的差值确定当前点的预测残差值；
    对所述最优预测模式及所述预测残差值进行编码。
  7. 根据权利要求5所述的方法，其特征在于，所述根据所述当前点的优选近邻点集进行编码处理，包括：
    将所述当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；
    依次将所述当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为所述点集中点的个数；
    计算所述预测模式1到所述预测模式N+1的代价值Jn，n由1到N+1，将所述Jn中的最小值确定为最小代价值，将所述最小代价值的对应模式确定为最优预测模式；
    根据所述最优预测模式确定当前点的预测值；
    根据所述当前点的属性值与所述当前点的预测值之间的差值确定当前点的预测残差值；
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，对所述最优预测模式及所述预测残差值进行编码；或，
    在所述第一距离与所述第二距离的差值大于所述设定值且所述第二代价值小于所述第一代价值时，对有效替换标志、所述最优预测模式及所述预测残差值进行编码，或，
    在所述第一距离与所述第二距离的差值大于所述设定值且所述第二代价值大于或等于所述第一代价值时，对无效替换标志、所述最优预测模式及所述预测残差值进行编码。
  8. 一种点云解码方法，其特征在于，包括：
    确定当前点的K个最近邻点组成的点集；
    确定所述当前点的L个次近邻点组成的点集；
    根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集；
    根据所述当前点的优选近邻点集进行解码处理。
  9. 根据权利要求8所述的方法，其特征在于，所述确定当前点的K个最近邻点组成的点集，包括：
    由点云中的点到所述当前点的空间距离大小确定所述当前点的K个最近邻点组成的点集。
    根据权利要求8所述的方法，其特征在于，所述确定当前点的L个次近邻点组成的点集，包括：
    由所述点云中的点到所述当前点的空间距离大小确定所述当前点的L个次近邻点组成的点集。
  10. 根据权利要求8至10所述的方法，其特征在于，所述根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值大于所述设定值时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  11. 根据权利要求8至10所述的方法，其特征在于，所述根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定当前点的优选近邻点集，包括：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值大于所述设定值且替换标志为有效替换标志时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值大于所述设定值且替换标志为无效替 换标志时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  12. 根据权利要求11中任一项所述的方法，其特征在于，所述根据所述当前点的优选近邻点集进行解码处理，包括：
    根据所述最优预测模式及所述优选近邻点集确定当前点的预测值；
    根据所述当前点的预测值与所述当前点的预测残差值的和确定所述当前点的属性值。
  13. 根据权利要求12中任一项所述的方法，其特征在于，所述根据所述当前点的优选近邻点集进行解码处理，包括：
    根据所述最优预测模式及所述优选近邻点集确定当前点的预测值；
    根据所述当前点的预测值与所述当前点的预测残差值的和确定所述当前点的属性值。
  14. 一种点云编码设备，其特征在于，包括：
    第一确定模块，用于确定当前点的K个最近邻点组成的点集；
    第二确定模块，用于确定所述当前点的L个次近邻点组成的点集；
    第三确定模块，用于根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定所述当前点的优选近邻点集；
    编码模块，用于根据所述当前点的优选近邻点集进行编码处理。
  15. 根据权利要求15所述的编码设备，其特征在于，所述第一确定模块具体用于：
    根据点云中的点到所述当前点的空间距离大小确定所述当前点的K个最近邻点组成的点集。
  16. 根据权利要求15所述的编码设备，其特征在于，所述第二确定模块具体用于：
    根据点云中的点到所述当前点的空间距离大小确定所述当前点的L个次近邻点组成的点集。
  17. 根据权利要求15至17所述的编码设备，其特征在于，所述第三确定模块具体用于：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L 个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值大于所述设定值时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  18. 根据权利要求15至17所述的编码设备，其特征在于，所述第三确定模块具体用于：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值大于所述设定值时：
    将所述当前点的K个最近邻点组成的点集的最小代价值确定为第一代价值；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的最小代价值确定为第二代价值；
    在所述第二代价值小于所述第一代价值时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第二代价值大于或等于所述第一代价值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  19. 根据权利要求18所述的编码设备，其特征在于，所述编码模块具体用于：
    将所述当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；
    依次将所述当前点的优选近邻点集中各点的重建属性值作为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为所述 点集中点的个数；
    计算所述预测模式1到所述预测模式N+1的代价值Jn，n由1到N+1，将所述Jn中的最小值确定为最小代价值，将所述最小代价值的对应模式确定为最优预测模式；
    根据所述最优预测模式确定当前点的预测值；
    根据所述当前点的属性值与所述当前点的预测值之间的差值确定当前点的预测残差值；
    对所述最优预测模式及所述预测残差值进行编码。
  20. 根据权利要求19所述的编码设备，其特征在于，所述编码模块具体用于：
    将所述当前点的优选近邻点集中各点的重建属性值的加权平均值确定为备选预测值1，将对应模式确定为预测模式1；
    依次将所述当前点的优选近邻点集中各点的重建属性值确定为备选预测值2到备选预测值N+1，依次将对应模式确定为预测模式2到预测模式N+1，其中N为所述点集中点的个数；
    计算所述预测模式1到所述预测模式N+1的代价值Jn，n由1到N+1，将所述Jn中的最小值确定为最小代价值，将所述最小代价值的对应模式确定为最优预测模式；
    根据所述最优预测模式确定当前点的预测值；
    根据所述当前点的属性值与所述当前点的预测值之间的差值确定当前点的预测残差值；
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，对所述最优预测模式及所述预测残差值进行编码，或，
    在所述第一距离与所述第二距离的差值大于所述设定值且所述第二代价值小于所述第一代价值时，对有效替换标志、所述最优预测模式及所述预测残差值进行编码，或，
    在所述第一距离与所述第二距离的差值大于所述设定值且所述第二代价值大于或等于所述第一代价值时，对无效替换标志、所述最优预测模式及所述预测残差值进行编码。
  21. 一种点云解码设备，其特征在于，包括：
    第一确定模块，用于确定当前点的K个最近邻点组成的点集；
    第二确定模块，用于确定所述当前点的L个次近邻点组成的点集；
    第三确定模块，用于根据所述当前点的K个最近邻点组成的点集及所述当前点的L个次近邻点组成的点集，确定所述当前点的优选近邻点集；
    解码模块，用于根据所述当前点的优选近邻点集进行解码处理。
  22. 根据权利要求22所述的解码设备，其特征在于，所述第一确定模块具体用于：
    根据点云中的点到所述当前点的空间距离大小确定所述当前点的K个最近邻点组成的点集。
  23. 根据权利要求22所述的解码设备，其特征在于，所述第二确定模块具体用于：
    根据点云中的点到所述当前点的空间距离大小确定所述当前点的L个次近邻点组成的点集。
  24. 根据权利要求22至24所述的解码设备，其特征在于，所述第三确定模块具体用于：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值大于所述设定值时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  25. 根据权利要求22至24所述的解码设备，其特征在于，所述第三确定模块具体用于：
    将所述当前点的K个最近邻点组成的点集的质心到所述当前点的距离确定为第一距离；
    将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集的质心到所述当前点的距离确定为第二距离；
    在所述第一距离与所述第二距离的差值小于或等于所述设定值时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值大于所述设定值且替换标志为有效替换标志时，将所述当前点的K个最近邻点组成的点集中K-H个最近邻点及所述当前点的L个次近邻点组成的点集中I个次近邻点共K-H+I个点组成的点集确定为所述当前点的优选近邻点集，或，
    在所述第一距离与所述第二距离的差值大于所述设定值且替换标志为无效替换标志时，将所述当前点的K个最近邻点组成的点集确定为所述当前点的优选近邻点集。
  26. 根据权利要求25中任一项所述的解码设备，其特征在于，所述解码模块具体用于：
    根据所述最优预测模式及所述优选近邻点集确定当前点的预测值；
    根据所述当前点的预测值与所述当前点的预测残差值的和确定所述当前点的属性值。
    根据权利要求26中任一项所述的解码设备，其特征在于，所述解码模块具体用于：
    根据所述最优预测模式及所述优选近邻点集确定当前点的预测值；
    根据所述当前点的预测值与所述当前点的预测残差值的和确定所述当前点的属性值。
PCT/CN2019/079980 2019-03-20 2019-03-28 点云编码和解码的方法、编码设备和解码设备 Ceased WO2020186548A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/279,177 US20220005230A1 (en) 2019-03-20 2019-03-28 Point Cloud Encoding and Decoding Methods, Encoding Device and Decoding Device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910214173.6 2019-03-20
CN201910214173.6A CN109889840B (zh) 2019-03-20 2019-03-20 点云编码和解码的方法、编码设备和解码设备

Publications (1)

Publication Number Publication Date
WO2020186548A1 true WO2020186548A1 (zh) 2020-09-24

Family

ID=66933316

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/079980 Ceased WO2020186548A1 (zh) 2019-03-20 2019-03-28 点云编码和解码的方法、编码设备和解码设备

Country Status (3)

Country Link
US (1) US20220005230A1 (zh)
CN (1) CN109889840B (zh)
WO (1) WO2020186548A1 (zh)

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112188199B (zh) * 2019-07-03 2024-05-10 腾讯美国有限责任公司 自适应点云属性编码的方法、装置、电子设备和存储介质
WO2021003726A1 (zh) 2019-07-10 2021-01-14 深圳市大疆创新科技有限公司 一种数据编码、数据解码方法、设备及存储介质
CN110349230A (zh) * 2019-07-15 2019-10-18 北京大学深圳研究生院 一种基于深度自编码器的点云几何压缩的方法
CN110418135B (zh) * 2019-08-05 2022-05-27 北京大学深圳研究生院 一种基于邻居的权重优化的点云帧内预测方法及设备
WO2021062736A1 (zh) * 2019-09-30 2021-04-08 Oppo广东移动通信有限公司 划分方法、编码器、解码器以及计算机存储介质
CN111145090B (zh) * 2019-11-29 2023-04-25 鹏城实验室 一种点云属性编码方法、解码方法、编码设备及解码设备
WO2021138785A1 (zh) 2020-01-06 2021-07-15 Oppo广东移动通信有限公司 划分方法、编码器、解码器及计算机存储介质
CN114902285A (zh) 2020-01-06 2022-08-12 Oppo广东移动通信有限公司 一种最近邻搜索方法及装置、设备、存储介质
WO2021138787A1 (zh) * 2020-01-06 2021-07-15 Oppo广东移动通信有限公司 一种帧内预测方法、编码器、解码器及存储介质
CN113453018B (zh) * 2020-03-25 2022-05-10 浙江大学 一种点云属性值预测方法和装置
JP7419151B2 (ja) * 2020-04-21 2024-01-22 株式会社東芝 サーバ装置、情報処理方法およびプログラム
CN113596466A (zh) * 2020-04-30 2021-11-02 Oppo广东移动通信有限公司 一种最近邻搜索方法、编码器、解码器及存储介质
WO2021248390A1 (zh) * 2020-06-10 2021-12-16 深圳市大疆创新科技有限公司 点云的排序方法和装置
CN111953998B (zh) * 2020-08-16 2022-11-11 西安电子科技大学 基于dct变换的点云属性编码及解码方法、装置及系统
KR20230074504A (ko) 2020-09-25 2023-05-30 광동 오포 모바일 텔레커뮤니케이션즈 코포레이션 리미티드 포인트 클라우드 인코딩 방법, 포인트 클라우드 디코딩 방법 및 관련 장치
WO2022062369A1 (zh) 2020-09-25 2022-03-31 Oppo广东移动通信有限公司 点云编解码方法与系统、及点云编码器与点云解码器
CN114598892B (zh) * 2020-12-07 2024-01-30 腾讯科技（深圳）有限公司 点云数据编码方法、解码方法、装置、设备及存储介质
CN116601944A (zh) * 2020-12-08 2023-08-15 Oppo广东移动通信有限公司 点云编解码方法、编码器、解码器及计算机存储介质
JP7678112B2 (ja) * 2020-12-22 2025-05-15 オッポ広東移動通信有限公司 点群の符号化方法、復号化方法、エンコーダ及びデコーダ
KR20230125782A (ko) * 2020-12-28 2023-08-29 광동 오포 모바일 텔레커뮤니케이션즈 코포레이션 리미티드 포인트 클라우드 인코딩 및 디코딩 방법, 시스템, 포인트 클라우드 인코더 및 포인트 클라우드 디코더
WO2022183611A1 (zh) * 2021-03-04 2022-09-09 Oppo广东移动通信有限公司 帧内预测方法及装置、编解码器、设备、存储介质
CN113096199B (zh) * 2021-03-12 2023-01-03 中山大学 一种基于莫顿码的点云属性预测方法、装置和介质
WO2022257143A1 (zh) * 2021-06-11 2022-12-15 Oppo广东移动通信有限公司 帧内预测、编解码方法及装置、编解码器、设备、介质
CN117581537A (zh) * 2021-07-02 2024-02-20 北京小米移动软件有限公司 用于对点云进行编码和解码的方法
CN117751574A (zh) * 2021-08-23 2024-03-22 Oppo广东移动通信有限公司 解码方法、编码方法、解码器以及编码器
CN121309850A (zh) * 2024-07-09 2026-01-09 维沃移动通信有限公司 点云解码方法、点云编码方法及相关设备

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105740798A (zh) * 2016-01-27 2016-07-06 西安理工大学 一种基于结构分析的点云场景物体识别方法
US20170368755A1 (en) * 2016-06-22 2017-12-28 Massachusetts Institute Of Technology Methods and Apparatus for 3D Printing of Point Cloud Data
CN109166160A (zh) * 2018-09-17 2019-01-08 华侨大学 一种采用图形预测的三维点云压缩方法

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101404061B (zh) * 2008-11-12 2012-04-25 山东理工大学 三维散乱点云拓扑近邻数据查询方法
CN101650838A (zh) * 2009-09-04 2010-02-17 浙江工业大学 基于重采样方法和仿射聚类算法的点云简化处理方法
CN104159105B (zh) * 2014-08-22 2017-06-16 河海大学 基于区域质心的h.264帧内预测模式快速判定方法
CN104850712B (zh) * 2015-05-27 2018-10-26 山东理工大学 实物表面采样数据拓扑邻域查询方法
CN105550368A (zh) * 2016-01-22 2016-05-04 浙江大学 一种高维数据的近似最近邻检索方法及检索系统
US10229533B2 (en) * 2016-11-03 2019-03-12 Mitsubishi Electric Research Laboratories, Inc. Methods and systems for fast resampling method and apparatus for point cloud data
CN106952339A (zh) * 2017-03-15 2017-07-14 重庆邮电大学 一种基于最优传输理论的点云采样方法
US10462485B2 (en) * 2017-09-06 2019-10-29 Apple Inc. Point cloud geometry compression
CN108198244B (zh) * 2017-12-20 2020-11-10 中国农业大学 一种苹果叶片点云精简方法及装置
CN108322742B (zh) * 2018-02-11 2019-08-16 北京大学深圳研究生院 一种基于帧内预测的点云属性压缩方法
CN109257604B (zh) * 2018-11-20 2020-11-27 山东大学 一种基于tmc3点云编码器的颜色属性编码方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105740798A (zh) * 2016-01-27 2016-07-06 西安理工大学 一种基于结构分析的点云场景物体识别方法
US20170368755A1 (en) * 2016-06-22 2017-12-28 Massachusetts Institute Of Technology Methods and Apparatus for 3D Printing of Point Cloud Data
CN109166160A (zh) * 2018-09-17 2019-01-08 华侨大学 一种采用图形预测的三维点云压缩方法

Also Published As

Publication number Publication date
CN109889840A (zh) 2019-06-14
CN109889840B (zh) 2022-11-22
US20220005230A1 (en) 2022-01-06

Similar Documents

Publication Publication Date Title
CN109889840B (zh) 点云编码和解码的方法、编码设备和解码设备
CN110572655B (zh) 一种基于邻居权重的参数选取和传递的点云属性编码和解码的方法及设备
CN110418135B (zh) 一种基于邻居的权重优化的点云帧内预测方法及设备
CN112218079B (zh) 一种基于空间顺序的点云分层方法、点云预测方法及设备
CN108322742B (zh) 一种基于帧内预测的点云属性压缩方法
Huang et al. Octree-Based Progressive Geometry Coding of Point Clouds.
JP7520223B2 (ja) 点群階層化方法、デコーダ、エンコーダ及び記憶媒体
CN111405281A (zh) 一种点云属性信息的编码方法、解码方法、存储介质及终端设备
US20190355152A1 (en) Point cloud attribute compression method based on kd tree and optimized graph transformation
CN109257604A (zh) 一种基于tmc3点云编码器的颜色属性编码方法
CN108335335A (zh) 一种基于增强图变换的点云属性压缩方法
WO2021023206A1 (zh) 基于邻居权重优化的点云属性预测、编码和解码方法及设备
EP4542965A1 (en) Point cloud encoding method, point cloud decoding method, and related device
CN113784125A (zh) 一种点云属性预测方法及其设备
CN115914650A (zh) 点云编解码方法、编码器、解码器及存储介质
CN116233468A (zh) 点云解码方法、点云编码方法、装置、设备、介质及产品
CN103487833A (zh) 一种三维地震体矢量量化压缩体绘制方法
US20240428463A1 (en) Systems and Methods for Improving Compression of 3D Textures Using Unmapped Pixels
Peng et al. Progressive geometry encoder using octree-based space partitioning
CN120958824A (zh) 点云编解码方法、装置、设备及存储介质
CN118923107A (zh) 索引确定方法、装置、解码器以及编码器
US20250022207A1 (en) Systems and Methods for Improving Compression of 3D Textures Using Unmapped Pixels
US20260113436A1 (en) Coding method, decoding method, bit stream, coder, decoder, and storage medium
Daribo et al. Point cloud compression for grid-pattern-based 3D scanning system
Park et al. Progressive compression of geometry information with smooth intermediate meshes

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19920029

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19920029

Country of ref document: EP

Kind code of ref document: A1