As an introduction, let's try to process the point cloud.
The point cloud file used as a sample is located examples/testdata/fragment.pcd.
The first step is to import cupoch.
We will name it cph for short and use it in the following.
import cupoch as cphBefore we get into the actual program, let's check if CUDA is available.
If you get True when executing the following function, CUDA is available.
cph.utility.is_cuda_available()
# True or FalseIf the check is True, try to read the point cloud file.
pointcloud = cph.io.read_point_cloud("examples/testdata/fragment.pcd")Once successfully loaded, try to display the loaded point cloud. Use visualizer as follows.
cph.visualization.draw_geometries([pointcloud])If the following window appears, you have succeeded.
Close the window and then downsample. Call the function that performs downsampling as follows.
downsampled = pointcloud.voxel_down_sample(0.05)Draw the point cloud after downsampling.
That concludes our introduction! There's plenty of other features you can use and experiment with!