This is how you can create a realistic looking visualisation of a future wind farm by super imposing two images, a photo and renders of turbines. An example is given here

Modern cameras and phones often store the position of where it was taken. If the photo contains this information, it will automatically be used, otherwise you will have to specify this yourself.
​
It is important that the position is correct, but don't worry if you are unsure. In step 2 you can see the position on a map and also change the position by clicking on the map
​

In this step you will see two things at the same time, the photo and the virtual terrain and turbines. Your task is to match the two by scaling, rotating and tilting the virtual scene. Reference points are needed to know that the alignment is correct. The following animations shows how to align using the three options with the horizon as a reference

This is basically all you need in order to align the virtual scene with the photo. The "Terrain transparency" function can help you make sure that the alignment is correct.

These animations make it looks easy because the horizon is a very clear reference point. This is not always the case but the tool supports adding more reference points, this is a bit more advanced and will be explained further down in its own section.

The application does not know what parts of the turbine is infront or behind the objects in the photo, and this needs to be corrected by hand. The following animation shows how this can be done. A transparency is added to the turbine layer to more easy find which parts should be removed

In order to align the two images you need to have reference points. A reference point is something that you can identify in both the virtual scene and the photo. This can be an easily identifiable feature in the terrain for example.

Sometimes the terrain is very uniform and such feature is not easily identifiable.

Then you can use other features, such as houses or trees instead. These do not exist in the virtual scene, and a marker needs to be added at the right place instead.

The following animations shows how a marker can be placed on top of a well known feature in Oslo, Holmenkollen

The map can be expanded to more easily find the right area. Notice the red "laser point" that appears when you hover in the map. It points to the same location in the virtual scene.

Once I have found Holmenkollen, I can press "alt" on my keyboard and left click with my mouse. A red circle appears both on the map and on the virtual scene. This is now a reference point for Holmenkollen. With this method you are able to use features that are not in the virtual scene to reference with the image. Note that multiple reference points can be placed

The "laser pointer" when hovering with the mouse in the map can also be used to verify the alignment of the two images.

If you have alignment, pointing at something in the map should also point to the same thing on the photo. Here are some examples

Note that the placement of the reference point will always be on the ground, since the virtual environment does not have any buildings, trees, etc in its model. Therefore, you should see the laser pointer ponting towards the base of a building for example