There are three different basic methods for capturing an object:

1. Parallel Recording
2. Centre Recording
3. Scale Recording

All three methods can be combined, when recording a complex object. Furthermore, the object that is being recorded can be completely displayed in the recordings or divided in vertical and/or horizontal segments.

1. Parallel Recording

When using the strategy “Parallel Recording” images are taken along an imaginary line parallel to the object, for example its house facade.

Depending on the height of the object and the target resolution the images are either only vertical segments or both vertical and horizontal segments. 

For the first option the object must be visible in the usable image area. The camera always performs parallel to the object and the image section changes in the same extent as you move yourself.

For the second option images are taken in two or more rows. This is called “Slewing”. So, the basic idea is that every new row covers another horizontal part of the imaginary line of the object, e.g., a house facade. Here the same rules apply as before plus every new row must overlap horizontally with the one before by at least 60%. 

2. Centre Recording

This strategy is particularly convenient for free-standing objects. The basic idea of centre recording is best explained with the example of a monument. All images focus on an imaginary centre line, the middle of the object. As the movement from picture to picture should not be more than 10° this results in at least 36 photos. 

Depending on the size of the object several circuits must be taken in the same manner as it is done with tall objects when doing parallel recording or the distance must be considered. However, this might have some effects on the target resolution. The distance between the pictures than will be calculated as given above. 

3. Scale Recording

The basic idea of scale recording is to optimize the target resolution for specific details on the object. For example, there might be a certain part like an oriel that needs a higher target resolution than the rest of the wall. However, the pictures from a closer perspective still must be connected by at least three points with the pictures that have a greater image mask in order to be properly referenced by the software. So, this strategy allows transitions from one scale to another. Here again the overlap should be about 80%.

4. Combined Recording

When facing complex objects the strategies might have to be combined to fullfil the necessary criteria. However, the best results will be achieved, when combining strategies object specifically. The most important criteria is to ensure that all pictures are connected with each other like a railwaynetwork in a toy train. Imagine the engine driver takes pictures of buildings while driving. 

Source: imgurl:http://www.modeltrainbooks.org/wp-content/uploads/2014/07/N-Scale-Model-Railroad-Layout-Image-2.jpg - Bing

Two examples may show how this will work.

a) Offset Recording

In this situation there are two facades, which are arranged in a 90° angle to each other. To connect the facades the camera must be tilted with a 90° angle. 

So, this requires the combination of the parallel and the centre recording strategy. First the facades are recorded parallel and in a second step, the two facades are connected by recording the corner via the centre strategy. Due to the 90° angle the recording must cover ¼ circuit, which would mean at least nine pictures. The distance orientates to the lines when recording parallel. 

In case of an internal corner the same procedure is applied. However, the corner must be recorded in a circuit surrounding the corner.

b) Complex Recording

When recording complex objects, like “L” or “U” shaped architecture, the same strategies will be applied. As in the given example all the lines are taken from the same distance. The lines, again, will be connected by a ¼ circle to connect them via the corner being the centre point. You might follow this scheme: