Constraints are processed by different parts of PhotoModeler depending on their type.
Multi-photo Constraints are processed only during Global Optimization.
Axes Constraints are processed by a special algorithm that processes just these single photo constraints and is run as part of Orientation. Axes Constraints may also be "reaffirmed" and processed after Global Optimization runs. This second stage provides the ability to use single photo techniques to complement a multi-photo project.
Control Points are used in both Orientation and Global Optimization.
Inverse Camera processing, described in the section Working with Unknown Cameras and Photographs, is done in the Orientation stage and can use both Axes Constraints and Control Points.
The Processing Dialog has a checkbox that allows the user to turn off all Constraint processing. This is useful for comparing results to non-constrained processing and for trouble shooting which is described in the next section.