Are there any ways to get a mean point cloud using frames from Azure recorded mkv?

### Checklist

- [X] I have searched for [similar issues](https://github.com/isl-org/Open3D/issues).
- [X] For Python issues, I have tested with the [latest development wheel](http://www.open3d.org/docs/latest/getting_started.html#development-version-pip).
- [X] I have checked the [release documentation](http://www.open3d.org/docs/release/) and the [latest documentation](http://www.open3d.org/docs/latest/) (for `master` branch).

### My Question

Hi,

Are there any ways to get a mean point cloud based on images extracted from Azure Kinect recorded mkv file (using `azure_kinect_mkv_reader`)? The number of pixels used with meaningful color and depth values from the extracted image series are slightly different. Thus, I can't directly average the point coordinates.

If not, are there any way to create mean color and depth image from the frames extracted from an mkv file? I tried to get a mean
depth array and then create a image using `img = o3d.geometry.Image((mean_depth).astype(np.uint8))` following this [example](http://www.open3d.org/docs/0.9.0/tutorial/Basic/working_with_numpy.html#from-numpy-to-open3d-image). However, np.asarray(img) because different from the original mean depth array.