Scanning in 3D things and objects from the real world attracted me since some time ago. After becoming aware of the techniques involved and doing some tests with the most affordable of them, it wasn’t until past year that I was able to use successfully a tool which used a photogrammetric approach for this purpose. It is the Python Photogrammetry Toolbox, PPT and I’m going to talk further about it in another post.
What is photogrammetry
Photogrammetry is a technique which is used since lots of time ago, and is used to extract metric and dimensional data from photographic imagery. Comparing the way the same feature is shown in one and the other photographic document is what makes us able to make such calculations. This technique is widely used in cartography. In this case, usually two aerial photos from the same area are used to calculate distances and construct a digital elevation model.
1. Creting the pointcloud
This is what we are going to do on the next post and in which PPT assists us. From the analysis of each feature detected in the photos, a point is created. Each one of them has the pixel color of the feature analysed, and together with others forms what we call pointcloud. Although we are going to see some weird functions and things here, actually this is the straightest part of the process since there are almost no chances to make it wrong.
2. Processing the point cloud
Pointclouds could be more or less dense, but usually they are something difficult to deal with by an usual 3D application, Blender included. For this specialized programs exist, for example Meshlab. Such programs allows us to work easily and fast with the huge amounts of data involved in pointclouds, and we can convert them to meshes besides many other operations. Then we can work with those meshes in Blender or any other 3D application.
This part is a little bit more complex, since we have more options to choose from and so we have more chances to make mistakes. In another post we are going to see a workflow that did the trick for me.
This is the first article from the three I plan to publish by now about this subject. Here you can read the second installment, where I explain in detail how to create a pointcloud using the Python Photogrammetry Toolbox.