A point cloud processing method based on distributed parameter coding
A processing method and a technology for distributing parameters, applied in image data processing, neural learning methods, instruments, etc., can solve problems such as poor parameter interpretation, large amount of calculation, and inability to guarantee Lipschitz continuity of point cloud data.
Active Publication Date: 2022-08-05
ZHEJIANG UNIV
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0009] Using distribution parameters to represent point cloud data can skip the step of discretization and obtain continuous data, while maintaining the global information and three-dimensional information of point cloud data including special points; however, it also has some disadvantages : First of all, it has a large amount of calculation, and it needs to train the neural network according to the current point cloud data; secondly, it cannot guarantee the Lipschitz continuity of the point cloud data in the parameter space; finally, the parameter interpretation used to represent the point cloud distribution Poor sex
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0028] In order to describe the present invention more specifically, the technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0029] The present invention is based on the point cloud processing method of distribution parameter coding, which can represent the dynamic distribution of point cloud data in real time, and can also compress the existing point cloud data. The method firstly imports the position data of the point cloud at the current moment Data[x,y ,z], assuming that the environmental random point cloud distribution at the current moment is Pr([x,y,z]), according to this probability distribution function, a certain amount of environmental random point cloud data is randomly generated, and the previously imported point cloud data is labeled 1. Label the point cloud data randomly generated from the environment with a label of 0; then train a neural network with a given struc...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a point cloud processing method based on distribution parameter coding. The method fits the distribution state of the point cloud by training a neural network, shows the probability density function of the current point cloud distribution, and uses the parameters at this time as the current point. The characteristics of cloud data successfully compress a large amount of point cloud data to the order of parameter space. At the same time, the present invention can take the point cloud parameters as the output of the loop unit in the neural network for the situation of continuously obtaining the point cloud, so as to predict the transformation trend of the point cloud data distribution with the change of the external environment.
Description
technical field [0001] The invention belongs to the technical field of point cloud data compression processing, and in particular relates to a point cloud processing method based on distribution parameter coding. Background technique [0002] With the rapid development of 3D data acquisition equipment, it has become more and more convenient and efficient to describe 3D objects and reproduce 3D scenes with 3D point clouds. 3D point cloud is a new data format used to record and represent the surface information of three-dimensional objects; point cloud data is a collection of a series of points in space, these points contain three-dimensional coordinates and one or more attribute information, such as color, Normal vector, reflection intensity, etc. Compared with other 3D data formats, point clouds have the advantages of convenient acquisition and simple processing, so they are widely used in various emerging fields, such as augmented reality (AR), autonomous driving, and 3D p...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T17/00G06N3/04G06N3/08
CPCG06T17/00G06N3/08G06N3/045
Inventor 吕攀钱广一李红杨国青吴朝晖
Owner ZHEJIANG UNIV