Search

KR-20260064643-A - METHOD AND APPARATUS FOR ENCODING/DECODING OF LiDAR POINT CLOUD

KR20260064643AKR 20260064643 AKR20260064643 AKR 20260064643AKR-20260064643-A

Abstract

A three-dimensional point cloud encoding method according to the present disclosure may include: a step of separating an input three-dimensional point cloud into a ground point cloud and an object point cloud; a step of deriving planar information for the ground point cloud; a step of generating a three-dimensional point cloud with transformed geometric information by performing a transformation on the three-dimensional point cloud based on the planar information; and a step of encoding the planar information and the three-dimensional point cloud with transformed geometric information to generate a bitstream.

Inventors

  • 장은영
  • 차지훈
  • 장의선
  • 리신
  • 오재영

Assignees

  • 한국전자통신연구원

Dates

Publication Date
20260507
Application Date
20251031
Priority Date
20241031

Claims (20)

  1. Step of separating the input 3D point cloud into a ground point cloud and an object point cloud; A step of deriving planar information for the above-mentioned ground point cloud; A step of generating a 3D point cloud with transformed geometric information by performing a transformation on the 3D point cloud based on the planar information; and A 3D point cloud encoding method comprising the step of encoding the 3D point cloud, in which the planar information and the geometric information are converted, to generate a bitstream.
  2. In paragraph 1, The above planar information is derived based on either a single planar model or a planar model having a slope, and Based on the above single-plane model, a single plane is derived, and A 3D point cloud encoding method in which a plane having a slope is derived when based on a plane model having the slope above.
  3. In paragraph 1, The above planar information is derived based on at least one model among a few planar models or a few planar models having a slope, and When based on the above few planar models, at least two planes are derived, and A 3D point cloud encoding method in which at least two planes having slopes are derived when based on a few plane models having the above slopes.
  4. In paragraph 1, The above plane information is derived by constructing a plane equation, and The above plane equation is a three-dimensional point cloud encoding method constructed by calculating the plane normal vector and the distance from the origin to the plane.
  5. In paragraph 1, The above plane equation is constructed based on eigenvalue information and eigenvector information, and A three-dimensional point cloud encoding method in which the above eigenvalue information and the above eigenvector information are derived using a principal component analysis technique.
  6. In paragraph 1, The above transformation is performed by replacing the z-axis coordinate value for each point constituting the three-dimensional point cloud with a first coordinate value, and A three-dimensional point cloud encoding method in which the first coordinate value is the distance value from each point to a plane.
  7. In paragraph 6, The above first coordinate value is a three-dimensional point cloud encoding method that is encoded according to either lossless encoding or lossy encoding.
  8. In paragraph 1, The above planar information is included in the header information used in g-pcc and signaled, and The above header information is a three-dimensional point cloud encoding method defined in a geometry parameter set.
  9. In paragraph 8, Signaling the first flag in the above header information, A 3D point cloud encoding method in which it is determined whether to perform inverse compensation for the z-coordinate of the 3D point cloud using the planar information derived based on the first flag.
  10. One or more transmitters/receivers; One or more memories; and Includes one or more processors, The above one or more processors are: Separate the input 3D point cloud into a ground point cloud and an object point cloud; Deriving planar information for the above ground point cloud; By performing a transformation on the 3D point cloud based on the above planar information, a 3D point cloud with transformed geometric information is generated; A point cloud encoding device configured to generate a bitstream by encoding a three-dimensional point cloud in which the above planar information and the above geometric information have been converted.
  11. A step of generating a 3D point cloud decrypted from a bitstream; A step of decoding planar information from a bitstream; and A 3D point cloud decoding method comprising the step of generating a restored 3D point cloud by performing an inverse transformation on the geometric information of the 3D point cloud decoded based on the planar information.
  12. In Paragraph 11, The above planar information is derived based on either a single planar model or a planar model having a slope, and Based on the above single-plane model, a single plane is derived, and A 3D point cloud decoding method in which a plane having a slope is derived when based on a plane model having the above slope.
  13. In Paragraph 11, The above planar information is derived based on at least one model among a few planar models or a few planar models having a slope, and When based on the above few planar models, at least two planes are derived, and A 3D point cloud decoding method in which at least two planes having slopes are derived when based on a few plane models having the above slopes.
  14. In Paragraph 11, The above plane information is derived by constructing a plane equation, and The above plane equation is a 3D point cloud decoding method constructed by calculating the plane normal vector and the distance from the origin to the plane.
  15. In Paragraph 11, The above plane equation is constructed based on eigenvalue information and eigenvector information, and A three-dimensional point cloud decoding method in which the above eigenvalue information and the above eigenvector information are derived using a principal component analysis technique.
  16. In Paragraph 11, The above inverse transformation is performed by replacing the z-axis coordinate value for each point constituting the decoded 3D point cloud with a second coordinate value, and A three-dimensional point cloud decoding method in which the second coordinate value is a value in which the distance from each point to a plane is compensated.
  17. In Paragraph 17, The above z-axis coordinate value is a three-dimensional point cloud decoding method that is decoded according to either lossless decoding or lossy decoding.
  18. In Paragraph 11, The above planar information is included in the header information used in g-pcc and signaled, and The above header information is a three-dimensional point cloud decoding method defined in a geometry parameter set.
  19. In Paragraph 18, In the above header information, the first flag is signaled, A 3D point cloud decoding method in which it is determined whether to perform compensation for the z-coordinate of the 3D point cloud using the planar information derived based on the first flag.
  20. One or more transmitters/receivers; One or more memories; and Includes one or more processors, The above one or more processors are: Generate a 3D point cloud decrypted from a bitstream; Decode planar information from a bitstream; A 3D point cloud decoding device configured to generate a restored 3D point cloud by performing an inverse transformation on the geometric information of the 3D point cloud decoded based on the planar information.

Description

Method and apparatus for encoding/decoding of LiDAR point cloud The present invention relates to a method for encoding/decoding a LiDAR point cloud and an apparatus for performing the same. More specifically, the present invention relates to a method and an apparatus for performing the same. Thanks to advancements in depth sensor technologies such as laser scanners, depth cameras, and LiDAR, it has become possible to acquire accurate three-dimensional information about objects and/or surrounding environments. The three-dimensional information acquired through these depth sensors is being actively utilized in immersive media service applications such as AR/VR or autonomous driving applications. Three-dimensional information acquired through a depth sensor can typically be represented as a three-dimensional point cloud. A three-dimensional point cloud may refer to a set of points representing an object and/or surrounding environment existing in three-dimensional space. A three-dimensional point cloud may include location information and/or attribute information in actual space. The attribute information may include color, reflectivity, etc. 3D point clouds have the problem of being larger in size compared to 2D images because they contain actual spatial information about objects. In addition, 3D point clouds are characterized by having different data properties compared to 2D images. Accordingly, various methods for compressing 3D point clouds are being studied. FIG. 1 is a block diagram of an encoding device that performs a point cloud encoding method using planar prediction according to one embodiment of the present disclosure. Figure 2 is a diagram showing an example of a point cloud acquired by a LiDAR sensor. FIG. 3 is a drawing showing an example of a separated point cloud according to one embodiment of the present disclosure. FIG. 4 is a drawing showing an example of a three-dimensional point cloud in which geometric information has been converted, according to the present disclosure. FIG. 5 is a block diagram of an apparatus for performing a planar information derivation method according to one embodiment of the present disclosure. FIG. 6 is a block diagram of a decoding device that performs a point cloud decoding method using planar prediction according to one embodiment of the present disclosure. FIG. 7 is a flowchart of a point cloud encoding method using planar prediction according to one embodiment of the present disclosure. FIG. 8 is a flowchart of a planar information derivation method according to one embodiment of the present disclosure. FIG. 9 is a flowchart of a point cloud decoding method using planar prediction according to one embodiment of the present disclosure. FIG. 10 is a block diagram illustrating a point cloud encoding device or a point cloud decoding device according to an embodiment of the present disclosure. The present disclosure is subject to various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present disclosure to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the present disclosure. Similar reference numerals in the drawings refer to the same or similar functions across various aspects. The shapes and sizes of elements in the drawings may be exaggerated for clearer explanation. The detailed description of exemplary embodiments described below refers to the accompanying drawings, which illustrate specific embodiments as examples. These embodiments are described in sufficient detail to enable those skilled in the art to practice the embodiments. It should be understood that various embodiments are different but need not be mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be implemented in other embodiments without departing from the spirit and scope of the present disclosure in relation to one embodiment. It should also be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the embodiment. Accordingly, the following detailed description is not intended to be taken in a limiting sense, and the scope of exemplary embodiments is limited only by the appended claims, together with all equivalents to those claimed therein, provided they are properly described. In this disclosure, terms such as first, second, etc. may be used to describe various components, but said components should not be limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of this disclosure, the first component may be named the second component, and similarly, the second component ma