add getting started

README.md

@@ -100,6 +100,10 @@ cmake -DBUILD_GLEW=ON -DBUILD_GLFW=ON -DBUILD_PNG=ON -DBUILD_JSONCPP=ON ..
 sudo make install-pip-package
 ```
 
+## Getting Started
+
+Please see how to use cupoch in [Getting Started](https://github.com/neka-nat/cupoch/blob/master/docs/getting_started.md) first.
+
 ## Results
 The figure shows Cupoch's point cloud algorithms speedup over Open3D.
 The environment tested on has the following specs:

docs/getting_started.md

@@ -0,0 +1,50 @@
+# Getting Started
+
+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.
+
+```py
+import cupoch as cph
+```
+
+Before 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.
+
+```py
+cph.utility.is_cuda_available()
+# True or False
+```
+
+If the check is `True`, try to read the point cloud file.
+
+```py
+pointcloud = cph.io.read_point_cloud("examples/testdata/fragment.pcd")
+```
+
+Once successfully loaded, try to display the loaded point cloud.
+Use visualizer as follows.
+
+```py
+cph.visualization.draw_geometries([pointcloud])
+```
+
+If the following window appears, you have succeeded.
+
+![getting_started_1](https://raw.githubusercontent.com/neka-nat/cupoch/master/docs/getting_started_1.png)
+
+Close the window and then downsample.
+Call the function that performs downsampling as follows.
+
+```py
+downsampled = pointcloud.voxel_down_sample(0.05)
+```
+
+Draw the point cloud after downsampling.
+
+![getting_started_2](https://raw.githubusercontent.com/neka-nat/cupoch/master/docs/getting_started_2.png)
+
+That concludes our introduction!
+There's plenty of other features you can use and experiment with!