IP Library › Granted Patent US 12,632,997
Granted Patent B2
US 12,632,997 · App. 18/015,835 · Granted May 19, 2026

Point cloud data transmission device, point cloud data transmission method, point cloud data reception device, and point cloud data reception method

Inventors: Sooyeon Lee (Seoul, KR); Sejin Oh (Seoul, KR); Yousun Park (Seoul, KR)
Assignee: LG ELECTRONICS INC.
G06T9/40G06T17/00G06T2210/56
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,632,997
App. No.
18/015,835
Granted
May 19, 2026
Kind
B2
Abstract

A point cloud data transmission method according to embodiments may comprise the steps of: acquiring point cloud data; encoding geometry information including the positions of points of the point cloud data; encoding, on the basis of the geometry information, attribute information including attribute values of the points of the point cloud data; and transmitting the encoded geometry information, the encoded attribute information, and signaling information.

Claims (58)

1 . A point cloud data encoding method, comprising:

acquiring point cloud data;

encoding geometry information including positions of points of the point cloud data based on an octree or a predictive tree;

encoding attribute information including attribute values of the points of the point cloud data based on the geometry information; and

transmitting the encoded geometry information, the encoded attribute information, and signaling information,

wherein the encoding the geometry information based on the predictive tree comprises:

generating the predictive tree based on the points of the point cloud data, and

compressing positions of points in the predictive tree by performing prediction on the positions of the points in the predictive tree based on one of a first prediction mode, a second prediction mode or a third prediction mode,

wherein traversal of the predictive tree is in depth-first order,

wherein the first prediction mode predicts a position of a current point based on a position of a parent node,

wherein the second prediction mode predicts the position of the current point based on a position derived by doubling the position of the parent node and subtracting a position of a grandparent node, and

wherein the third prediction mode predicts the position of the current point based on a position derived by adding the position of the parent node and the position of the grandparent node, and subtracting a position of a great-grandparent node.

2 . The point cloud data encoding method of claim 1 , wherein the generating the predictive tree further includes determining at least one inferred direct coding mode (IDCM) point to which a direct coding mode is to be applied from the predictive tree.

3 . The point cloud data encoding method of claim 2 , wherein the determining the IDCM includes determining the current point as the at least one IDCM point based on a difference between an average value of distances of points in a parent and child relationship from a parent node of the current point to a root node while the predictive tree is being generated and a distance between the current point and a parent point of the current point being greater than a preset threshold.

4 . The point cloud data encoding method of claim 2 , wherein the determining the IDCM includes positioning the at least one IDCM point at a leaf node based on the at least one IDCM point being positioned between a root node and the leaf node in the predictive tree.

5 . The point cloud data encoding method of claim 2 , wherein the determining the IDCM includes calculating an imaginary value to be allocated to the determined at least one IDCM point and allocating the imaginary value to the determined at least one IDCM point.

6 . A point cloud data encoding device, comprising:

an acquirer configured to acquire point cloud data;

a geometry encoder configured to encode geometry information including positions of points of the point cloud data based on an octree or a predictive tree;

an attribute encoder configured to encode attribute information including attribute values of the points of the point cloud data based on the geometry information; and

a transmitter configured to transmit the encoded geometry information, the encoded attribute information, and signaling information,

wherein the geometry encoder based on the predictive tree includes:

a predictive tree generator configured to generate the predictive tree based on the points of the point cloud data, and

a compressor configured to compress positions of points in the predictive tree by performing prediction on the positions of the points in the predictive tree based on one of a first prediction mode, a second prediction mode or a third prediction mode,

wherein traversal of the predictive tree is in depth-first order,

wherein the first prediction mode predicts a position of a current point based on a position of a parent node,

wherein the second prediction mode predicts the position of the current point based on a position derived by doubling the position of the parent node and subtracting a position of a grandparent node, and

wherein the third prediction mode predicts the position of the current point based on a position derived by adding the position of the parent node and the position of the grandparent node, and subtracting a position of a great-grandparent node.

7 . The point cloud data encoding device of claim 6 , wherein the predictive tree generator determines at least one inferred direct coding mode (IDCM) point to which a direct coding mode is to be applied from the predictive tree.

8 . The point cloud data encoding device of claim 7 , wherein the predictive tree generator determines the current point as the at least one IDCM point based on a difference between an average value of distances of points in a parent and child relationship from a parent node of the current point to a root node while the predictive tree is being generated and a distance between the current point and a parent point of the current point being greater than a preset threshold.

9 . The point cloud data encoding device of claim 7 , wherein the predictive tree generator positions the at least one IDCM point at a leaf node based on the at least one IDCM point being positioned between a root node and the leaf node in the predictive tree.

10 . The point cloud data encoding device of claim 7 , wherein the predictive tree generator calculates an imaginary value to be allocated to the determined at least one IDCM point and allocates the imaginary value to the determined at least one IDCM point.

11 . A point cloud data decoding method, comprising:

receiving geometry information, attribute information, and signaling information;

decoding the geometry information based on an octree or a predictive tree to derive positions of points; and

decoding the attribute information based on the signaling information and the geometry information to derive attribute values of the points,

wherein the decoding the geometry information based on the predictive tree derives the positions of the points by performing prediction in the predictive tree based on one of a first prediction mode, a second prediction mode or a third prediction mode,

wherein traversal of the predictive tree is in depth-first order,

wherein the first prediction mode predicts a position of a current point based on a position of a parent node,

wherein the second prediction mode predicts the position of the current point based on a position derived by doubling the position of the parent node and subtracting a position of a grandparent node, and

wherein the third prediction mode predicts the position of the current point based on a position derived by adding the position of the parent node and the position of the grandparent node, and subtracting a position of a great-grandparent node.

12 . The point cloud data decoding method of claim 11 , wherein the decoding the geometry information includes identifying at least one inferred direct coding mode (IDCM) point to which a direct coding mode is to be applied among the points based on the signaling information.

13 . The point cloud data decoding method of claim 12 , wherein the decoding the geometry information includes restoring the at least one IDCM point to an original point based on the signaling information.

14 . The point cloud data decoding method of claim 12 , wherein the decoding the geometry information includes performing prediction based on an imaginary value allocated to a position of the at least one IDCM point based on the signaling information.

15 . The point cloud data decoding method of claim 11 , wherein the signaling information includes at least one of IDCM determination-related information, imaginary value-related information, threshold value-related information, or index-related information.

16 . A point cloud data decoding device, comprising:

a receiver configured to receive geometry information, attribute information, and signaling information;

a geometry decoder configured to decode the geometry information based on an octree or a predictive tree to derive positions of points; and

an attribute decoder configured to decode the attribute information based on the signaling information and the geometry information to derive attribute values of the points,

wherein the geometry decoder based on the predictive tree derives the positions of the points by performing prediction in the predictive tree based on one of a first prediction mode, a second prediction mode or a third prediction mode,

wherein traversal of the predictive tree is in depth-first order,

wherein the first prediction mode predicts a position of a current point based on a position of a parent node,

wherein the second prediction mode predicts the position of the current point based on a position derived by doubling the position of the parent node and subtracting a position of a grandparent node, and

wherein the third prediction mode predicts the position of the current point based on a position derived by adding the position of the parent node and the position of the grandparent node, and subtracting a position of a great-grandparent node.

17 . The point cloud data decoding device of claim 16 , wherein the geometry decoder identifies at least one inferred direct coding mode (IDCM) point to which a direct coding mode is to be applied among the points based on the signaling information.

18 . The point cloud data decoding device of claim 17 , wherein the geometry decoder restores the at least one IDCM point to an original point based on the signaling information.

19 . The point cloud data decoding device of claim 17 , wherein the geometry decoder performs prediction based on an imaginary value allocated to a position of the at least one IDCM point based on the signaling information.

20 . The point cloud data decoding device of claim 16 , wherein the signaling information includes at least one of IDCM determination-related information, imaginary value-related information, threshold value-related information, or index-related information.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 12, 2023
From: LEE, SOOYEON; OH, SEJIN; PARK, YOUSUN
To: LG ELECTRONICS INC.
Reel/Frame 062360/0099 →
Priority Claims (1)
KR 10-2020-0085870 · Jul 13, 2020 · national
Continuity (1)
Related Publication 20230260163A1 · Aug 17, 2023
References Cited (13)
US 10609423B2 · Chuang · 2020 [cited by examiner]
US 20200107048A1 · Yea · 2020 [cited by examiner]
US 20210218994A1 · Flynn · 2021 [cited by examiner]
US 20210312670A1 · Mammou · 2021 [cited by examiner]
KR 1020190044629 · 2019 [cited by applicant]
Zhang et al., “Implicit Geometry Partition for Point Cloud Compression”, 2020 (Year: 2020). [cited by examiner]
Li et al., “Advanced 3D Motion Prediction for Video-Based Dynamic Point Cloud Compression”, (Year: 2020). [cited by examiner]
Lu et al., “MoNet: Motion-Based Point Cloud Prediction Network”, (Year: 2022). [cited by examiner]
Li et al., “Motion Estimation and Coding Structure for Inter-Prediction of LiDAR Point Cloud”, (Year: 2021). [cited by examiner]
Mammou, Khaled et al., “G-PCC codec description v2”, International Organisation for Standardisation Organisation Internationale De Normalisation ISO/IEC JTC1/SC29/WG11 Coding of Moving Pictures and Audio. N18189. Jan. 2… [cited by applicant]
Gao, Zhen et al., “[G-PCC] [New proposal] Predictive Geometry Coding”, International Organisation for Standardisation Organisation Internationale De Normalisation ISO/IEC JTC1/SC29/WG11 Coding of Moving Pictures and Aud… [cited by applicant]
Flynn, David, et al., “G-PCC: Combined octree-predictive geometry coding”, International Organisation for Standardisation Organisation Internationale De Normalisation ISO/IEC JTC1/SC29/WG11 Coding of Moving Pictures and… [cited by applicant]
Flynn, David, et al., “G-PCC: An IDCM specific QP for in-tree geometry quantisation”, International Organisation for Standardisation Organisation Internationale De Normalisation ISO/IEC JTCI/SC29AVG11 Coding of Moving P… [cited by applicant]