【点云可视化】自制ply文件并使用open3d可视化点云

使用open3d可视化点云，需要将点云制作为ply文件传入函数中。

安装open3d

直接使用pip安装

pip install open3d

ply文件

ply文件的结构为：文件头+数据内容，文件头从 ply 一行开始，到 end_header 为止，里面定义了此 ply文件中数据的组织形式，包括数据类型和个数。后面是点云的坐标信息及每个点的颜色属性的，参考自

如下为具有5个点的点云，end_header后每一行表示一个点的x,y,z,R,G,B信息

ply
format ascii 1.0
comment PCL generated
element vertex 5
property float x
property float y
property float z
property uchar red
property uchar green
property uchar blue
end_header
-0.25412 -0.11759 0.01171 255 0 255
0.18866 0.0793 0.00609 255 0 255
0.03777 0.06232 -0.00753 255 0 255
-0.25185 -0.08387 -0.01735 255 0 255
-0.35174 -0.16687 -0.00875 0 255 0

假设该文件为test.ply文件，则pyhon中使用open3d进行可视化命令为

import open3d as o3d
# visualization of point clouds.
pcd = o3d.io.read_point_cloud('test.ply')
o3d.visualization.draw_geometries([pcd])

自制ply文件

将numpy格式的点云转换为ply文件进行可视化很简单，根据上面的格式来写文件即可，下面给出一个参考

import open3d as o3d
import numpy as np

def makePlyFile(xyzs, labels, fileName='makeply.ply'):
    '''Make a ply file for open3d.visualization.draw_geometries
    :param xyzs:    numpy array of point clouds 3D coordinate, shape (numpoints, 3).
    :param labels:  numpy array of point label, shape (numpoints, ).
    '''
    RGBS = [
        (0, 255, 255),
        (255, 0, 0),
        (0, 255, 0),
        (0, 0, 255),
        (0, 0, 0),
        (255, 0, 255)
    ]

    with open(fileName, 'w') as f:
        f.write('ply\n')
        f.write('format ascii 1.0\n')
        f.write('comment PCL generated\n')
        f.write('element vertex {}\n'.format(len(xyzs)))
        f.write('property float x\n')
        f.write('property float y\n')
        f.write('property float z\n')
        f.write('property uchar red\n')
        f.write('property uchar green\n')
        f.write('property uchar blue\n')
        f.write('end_header\n')
        for i in range(len(xyzs)):
            r, g, b = RGBS[int(labels[i])-1]
            x, y, z = xyzs[i]
            f.write('{} {} {} {} {} {}\n'.format(x, y, z, r, g, b))

if __name__ == "__main__":
    numpoints = 2000  # 2000个点
    xyzs = np.random.rand(numpoints, 3)
    labels = np.random.randint(0, 4, numpoints)  # 4种label
    makePlyFile(xyzs, labels, 'demo.ply')
    pcd = o3d.io.read_point_cloud('demo.ply')
    o3d.visualization.draw_geometries([pcd])