US 12,277,647 B2
Method for processing three-dimensional data and device
Yunkai Du, Shenzhen (CN); Yilong Chen, Shenzhen (CN); and Yang Yang, Beijing (CN)
Assigned to HUAWEI TECHNOLOGIES CO., LTD., Guangdong (CN)
Filed by HUAWEI TECHNOLOGIES CO., LTD., Guangdong (CN)
Filed on Mar. 22, 2023, as Appl. No. 18/187,771.
Application 18/187,771 is a continuation of application No. PCT/CN2021/096589, filed on May 28, 2021.
Claims priority of application No. 202011010600.8 (CN), filed on Sep. 23, 2020.
Prior Publication US 2023/0222732 A1, Jul. 13, 2023
Int. Cl. G06T 17/00 (2006.01); G06T 5/20 (2006.01)
CPC G06T 17/00 (2013.01) [G06T 5/20 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for processing three-dimensional data, comprising:
inputting first data, at least one first background value of the first data, and a first sparse distribution diagram of the first data into a target convolution operator of a convolutional neural network, wherein the convolutional neural network comprises at least one first convolution operator, and the target convolution operator is any one of the at least one first convolution operator, wherein the first data comprises a plurality of first voxels, the at least one first background value indicates at least one background in the first data, and the at least one first background value is in a one-to-one correspondence with the at least one background, and wherein the first sparse distribution diagram describes a sparse distribution of the plurality of first voxels by using a background identifier and a non-background identifier, the background identifier indicates that a first voxel that corresponds to the background identifier in the first data is a background, and the non-background identifier indicates that a first voxel that corresponds to the non-background identifier in the first data is a non-background;
searching, based on a background identifier of a background point that corresponds to a first voxel indicating a background in the first data in the first sparse distribution diagram, a correspondence for an intermediate result corresponding to the background identifier of the background point, wherein the correspondence indicates a relationship between an intermediate result of each of the at least one first background value and a background identifier corresponding to each of the at least one first background value, the intermediate result of each first background value in the at least one first background value is obtained through calculation by using the first background value and a convolution kernel of the target convolution operator, and the intermediate result corresponding to the background identifier of the background point is used to determine a value of a second voxel in second data; and
controlling the target convolution operator to output the second data, wherein the second data comprises a plurality of second voxels.