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

1 . A method of transmitting point cloud data by an encoder, the method comprising:

encoding point cloud data; and

transmitting a bitstream containing the point cloud data;

wherein the encoding of the point cloud data comprises:

encoding geometry information of the point cloud data;

encoding attribute information of the point cloud data,

wherein the encoding the attribute information includes:

generating levels of detail (LoDs) for the attribute information,

wherein a quantization parameter is applied to the attribute information,

wherein the attribute information is predicted based on count information,

wherein the encoding of the attribute information of the point cloud data comprises:

generating near neighbor point information for each of points of the point cloud data based on reconstructed geometry information;

calculating a prediction reference count for each of the points based on the near neighbor point information; and

calculating the quantization parameter for each of the points based on the prediction reference count.

2 . The method of claim 1 ,

wherein the quantization parameter is related to the near neighbor point information,

wherein the attribute information is quantized based on the quantization parameter.

3 . The method of claim 2 , the method further comprising:

calculating an influence weight for each of the points based on the near neighbor point information; and

calculating the quantization parameter for each of the points based on the influence weight.

4 . The method of claim 2 , the method further comprising:

calculating a threshold or a delta quantization parameter based on the prediction reference count; and

calculating the quantization parameter based on the threshold and the delta quantization parameter.

5 . The method of claim 1 , wherein encoding of the attribute information of the point cloud data comprises:

calculating a delta quantization parameter for the points based on a layer of the LoD; and

calculating a quantization parameter based on the delta quantization parameter.

6 . A method comprising:

encoding geometry information of point cloud data,

encoding attribute information of the point cloud data, and

wherein the encoding the attribute information includes:

generating a level of levels of detail (LoDs);

obtaining neighboring points for a point in the level;

predicting attribute information for the point based on the neighboring points;

quantizing residual information for the point based on a quantization weight of the point; and

reconstructing attribute information for the point based on the predicted attribute information and the scaled residual attribute information, and

wherein the quantization weight of the point is derived based on whether the point is used to predict attribute information for a second point of LoDs.

7 . The method of claim 6 , wherein the encoding the attribute information further includes:

generating near neighbor point information for the point based on reconstructed geometry information; and

calculating the quantization weight based on the near neighbor point information.

8 . The method of claim 7 , wherein the encoding the attribute information further includes:

calculating a threshold or a delta quantization parameter based on the quantization weight; and

calculating the quantization parameter based on the threshold and the delta quantization parameter.

9 . A method comprising:

decoding geometry information of point cloud data; and

decoding attribute information of the point cloud data using levels of detail (LoDs),

wherein the decoding the attribute information includes:

generating a level of the LoDs;

obtaining neighboring points for a point in the level;

predicting attribute information for the point based on the neighboring points;

scaling residual information for the point based on a quantization weight of the point; and

reconstructing attribute information for the point based on the predicted attribute information and the scaled residual attribute information, and

wherein the quantization weight of the point is derived based on whether the point is used to predict attribute information for a second point of LoDs.

10 . The method of claim 9 ,

wherein the decoding the attribute information further includes:

obtaining near neighbor point information for the point, and

wherein the quantization weight of the point is obtained based on the near neighbor point information.

11 . The method of claim 9 , wherein the decoding of the attribute information of the point cloud data comprises:

obtaining a quantization parameter of the point based on a base quantization parameter and the quantization parameter offset of the point, and

wherein the base quantization parameter is obtained from a bitstream.

12 . A device comprising:

a memory; and

a processor connected to the memory, wherein the processor is configured to:

decode geometry information of point cloud data; and

decode attribute information of the point cloud data using levels of detail (LoDs),

wherein, to decode the geometry information, the processor is further configured to:

generate a level of the LoDs;

obtain neighboring points for a point of the level:

predict attribute information for the point based on the neighboring points;

scale residual information for the point based on a quantization weight of the point; and

reconstruct attribute information for the point based on the predicted attribute information and the scaled residual attribute information, and

wherein the quantization weight of the point is derived based on whether the point is used to predict attribute information for a second point of LoDs.

13 . The device of claim 12 ,

wherein the processor is further configured to:

obtain near neighbor point information for the point, and

wherein the quantization weight of the point is obtained based on the near neighbor point information.

14 . The device of claim 12 , wherein the processor is further configured to:

obtain a quantization parameter of the point based on a base quantization parameter and the quantization parameter offset of the point, and

wherein the base quantization parameter is obtained from a bitstream.

15 . The device of claim 12 , wherein the processor is further configured to:

obtain a threshold for the quantization weight and a base quantization parameter from a bitstream;

based on the threshold and the quantization weight of the point, obtain a quantization parameter of the point based on the base quantization parameter and the quantization parameter offset of the point.