# US 11,122,293 B2 · Intra-frame prediction-based point cloud attribute compression method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,122,293 B2
- Document ID: US11122293B2
- Status: Granted
- Kind: B2
- Application number: 16/955,615
- Filed: 2018-02-12
- Published: 2020-11-19
- Granted: 2021-09-14
- Earliest claimed: 2018-02-11
- Assignees: Peking University Shenzhen Graduate School
- Applicants: Peking University Shenzhen Graduate School
- Inventors: Ge Li; Yiting Shao
- Web page: https://patentgrep.com/patent/US11122293B2

## CPC

- H04N 19/593: involving spatial prediction techniques
- H04N 19/129: Scanning of coding units, e.g. zig-zag scan of transform coefficients or flexible macroblock ordering [FMO]
- H04N 19/136: Incoming video signal characteristics or properties
- H04N 19/156: Availability of hardware or computational resources, e.g. encoding based on power-saving criteria
- H04N 19/159: Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
- H04N 19/176: the region being a block, e.g. a macroblock
- H04N 19/186: the unit being a colour or a chrominance component
- H04N 19/61: in combination with predictive coding
- H04N 19/625: using discrete cosine transform [DCT]
- H04N 19/91: Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
- H04N 19/96: Tree coding, e.g. quad-tree coding
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

An intra-frame prediction-based point cloud attribute compression method. A new block structure-based intra-frame prediction scheme is provided for point cloud attribute information, where four prediction modes are provided to reduce information redundancy among different coding blocks as much as possible and improve point cloud attribute compression performance. The method comprises: performing point cloud attribute color space conversion; dividing a point cloud by using a K-dimensional (KD) tree to obtain coding blocks; performing block structure-based intra-frame prediction; performing intra-frame prediction mode division; performing conversion, uniform quantization, and entropy coding. The method relates to performing coding block division on a point cloud by using a KD tree, numbering coding blocks in a breadth-first traversal order, processing the coding blocks one by one in the numbering order by using a point cloud intra-frame prediction scheme having four prediction modes, selecting an optical prediction mode by means of mode decision, and performing conversion, quantization, and entropy coding on a predicted residual, so as to achieve optimal performance of point cloud attribute compression.

## Claims

### Claim 1

An intra-frame prediction-based point cloud attribute compression method, wherein a new block structure-based intra-frame prediction scheme is provided for point cloud attribute information, four prediction modes are provided to reduce information redundancy among different coding blocks as much as possible to improve point cloud attribute compression performance; the method comprises the steps of:
1) color space transformation of point cloud attributes:
reading in point cloud attribute information to be processed, and transforming a point cloud color space from an RGB space to a YUV space according to the visual characteristics of human eyes and the difficulty degree of compression processing;
2) dividing the point cloud by using a KD tree to obtain coding blocks, and numbering the coding blocks according to a breadth traversal sequence
reading in geometric information of the point cloud, performing KD tree division on the point cloud according to the geometric information, selecting a coordinate axis with a largest distribution variance in the position coordinate of the point cloud as a division axis each time, selecting a point with the coordinate being a median value as a division point, and performing iterative division until reaching a set KD tree depth; a block obtained by a last layer from the KD tree division being a coding block of the point cloud, numbering the coding blocks according to a breadth traversal order, and the number being used as a sequence for later processing of the coding blocks;
3) performing intra-frame prediction on the attribute information of the point cloud coding block based on the numbering sequence, with four prediction modes:
when the intra-frame prediction is performed on a first coding block b1, 128 is used as a reference value for Y-component prediction; when color components Yi, Ui, and Vi of other blocks bi (i≠1) are processed, 128 is used as a prediction reference value of Yi in a prediction mode I, and Ui and Vi are not predicted; a mean value Yi-1 of Y-components of all points after reconstruction of a previous block bi-1 is used for predicting Yi in the prediction mode II, and Ui and Vi are not predicted; mean values Ui-1 and Vi-1 of the U and V color components of all points after the reconstruction of the previous block are used for predicting Ui and Vi in the prediction mode III, and Yi is not predicted; Yi-1, Ui-1, and Vi-1 are used as reference values of Yi, Ui and Vi for predicting in the prediction mode IV; and the prediction modes comprises four prediction modes;
4) mode decision for intra-frame prediction:
four modes are used for predicting color components Yi, Ui, and Vi of the coding block bi (i≠1), an optimal prediction mode is required to be selected by mode decision which is not required for a first coding block; estimating a cost of a prediction mode using a Sum of Absolute Transformed Differences (SATD) and an absolute value of a prediction residual transformation coefficient, wherein the prediction residual is transformed using a Discrete Cosine Transform (DCT); a smaller SATD value represents a lower cost of the prediction mode and better prediction performance, and the mode with the smallest SATD is selected as the prediction mode of the current block; and
5) generation of a code stream with point cloud attribute compression: processing all the coding blocks according to the coding sequence, performing DCT transformation, uniform quantization and entropy coding on the predicted residual to obtain a final code stream with point cloud attribute compression.

### Claim 2 (depends on 1)

The point cloud attribute compression method of claim 1, wherein the specific process of the color space transformation in the step 1) is as follows: a point pi in the point cloud has color values ri, gi and bi in an RGB color space; the RGB is transformed into a YUV color space by a formula 1, and the color values thereof are yi, ui and vi.
[
y
i
,
u
i
,
v
i
]
=
[
r
i
,
g
i
,
b
i
]
×
[
0
.
2
⁢
1
⁢
2
⁢
6
0
.
7
⁢
1
⁢
5
⁢
2
0
.
0
⁢
7
⁢
2
⁢
2
-
0
.
1
⁢
1
⁢
4
⁢
6
-
0
.
3
⁢
8
⁢
5
⁢
4
0
.
5
0
.
5
-
0
.
4
⁢
5
⁢
4
⁢
2
-
0
.
0
⁢
4
⁢
5
⁢
8
]
.
(
Formula
⁢
⁢
1
)

### Claim 3 (depends on 1)

The point cloud attribute compression method of claim 1, wherein the KD tree division method in the step 2) is a binary division method; let the point cloud to be processed have N points, and the division depth set by the KD tree be d, 2d coding blocks are obtained after the point cloud is divided for d times; and all coding blocks are numbered with b1, b2, . . . , b1, . . . , b2d in the breadth traversal order.

### Claim 4 (depends on 1)

The point cloud attribute compression method of claim 1, wherein a reference value used for performing intra-frame prediction on the coding block in the step 3) is shown in the following table 1:
TABLE 1
Color component reference values for
intra-frame prediction of coding blocks
Yi
Ui
Vi
Coding
Prediction
reference
reference
reference
block bi
mode
value
value
value
i = 1
1
128
0
0
i ≠ 1
1
128
0
0
2
Yi − 1
0
0
3
0
Ui − 1
Vi − 1
4
Yi − 1
Ui − 1
Vi − 1.

### Claim 5 (depends on 1)

The point cloud attribute compression method of claim 1, wherein the prediction reference values of the color components Yi, Ui and Vi of the coding block bi (i≠1) in the step 4) are Yi_ref, Ui_ref and Vi_ref, respectively, the prediction residual bi(res) is calculated by a formula 2, and the cost SATD of the prediction mode is calculated by a formula 3, with the specific process as follows:
bi(res)=(Yi−Yi_ref)+(Ui−Ui_ref)+(Vi−Vi_ref) (Formula 2);
SATD=sum(abs(DCT(bi(res)))) (Formula 3).

### Claim 6 (depends on 1)

The point cloud attribute compression method of claim 1, wherein the specific details in the step 5) comprise the steps of:
(6-1) performing the processes in the steps 1) to 4) to obtain numbered coding block prediction residuals and prediction mode information thereof; performing DCT transformation, uniform quantization and binarization on the prediction residuals to obtain a code stream with attribute information, combining the prediction mode information of the encoding block, and performing entropy encoding to obtain a final code stream with point cloud attribute compression;
(6-2) code stream of point cloud attribute information: the code stream of point cloud attribute information is mainly composed of compressed header information and coded block information; wherein the header information mainly comprises quantization step length and the like; the coded block information stream is arranged in the order of coded blocks in units of coded blocks, and each block mainly comprises prediction mode information and color residual information of the coding blocks; and
(6-3) the performance of point cloud attribute compression is measured by a bitrate and a Peak Signal to Noise Ratio (PSNR), wherein the ode rate is in units of bits per point (bpp), and PSNR is in decibel dB; the smaller the ode rate is, the larger the PSNR is, and the better the performance of point cloud attribute compression is.