Convert a pointcloud from .txt to .ply

Question

I am new to Pointcloud (LiDAR) processing and I was wondering if there is any tool/library able to convert a pointcloud saved in .txt to a .ply file.

The pointcloud .txt looks like this

pointcloud_width pointcloud_height
point1_x point1_y point1_z point1_intensity
point2_x point2_y point2_z point2_intensity
...
pointN_x pointN_y pointN_z pointN_intensity

Thanks for your help

Answer 1

You can use Open3d for this:

import open3d as o3d

pcd = o3d.io.read_point_cloud("input.txt", format="xyz")
o3d.io.write_point_cloud("output.ply", pcd)

This will not preserve the intensity values, though. If you need them use open3d.t:

pcd = o3d.t.io.read_point_cloud("input.txt", format="xyzi")
o3d.t.io.write_point_cloud("output.ply", pcd)

Note that PLY comes in two flavours: ASCII and binary. The above solution gives you a binary file, which is the default for open3d. If you really need ASCII then use this:

o3d.t.io.write_point_cloud("output.ply", pcd, write_ascii=True)

Alternatively, since ASCII is basically plain text, you could do this without any external library by just replacing the first line of your .txt file with a custom header so it looks like this:

ply
format ascii 1.0
element vertex {insert number of points here}
property float x
property float y
property float z
property float scalar_Intensity
end_header
point1_x point1_y point1_z point1_intensity
point2_x point2_y point2_z point2_intensity
...
pointN_x pointN_y pointN_z pointN_intensity

and saving it with a .ply extension.

Lastly, if you just need a "tool" then check out CloudCompare, which handles many file format conversions.