About PointCloud data from velodyne and RGB-D camera
Hello, I'm using ROS kinetic with Ubuntu 16.04.
I've addressed 2D-LiDAR's laser data for machine learning neural networks.
But I also think it might possible to bring 3D point cloud data to machine learning.
The 3D point cloud depth data from RGB-D camera (the type of topic message is sensor_msgs/PointCloud2.) has data part of 1D simple list data; it looks like just a simple list type data of Python.
Unfortunately, I have no idea how to deal with this data, I really don't know what the values of the 1D-list mean.
I guess it certainly has something to do with depth from the camera frame.
On the other hand, I've ever heard that PCL can help handle PointCloud data.
The only what I want to do is to convert raw PointCloud data of 1D to 3D-convolution data; so it needs to get (16,720) size data , for example.
Is there anyone who is familiar with this kind of transformation?; or anyone who knows what the raw 1D list of PointCloud data from RGB-D camera means? Just so you know, I'm working with 'Xtion Pro Live' camera of Asus.
Thanks in advance :)
