One of ContextCapture e's breakthrough features is its ability to handle photographs that have no positioning data. In such a case, ContextCapture generates a 3D model with arbitrary position, rotation and scale, yet with a plausible up vector. However, ContextCapture also natively supports several types of positioning data including GPS tags, control points and can import potentially any other positioning data through position/rotation import or complete block import.

GPS tags, if present in Exif metadata or in an accompanying XMP file, are automatically extracted, and can be used to georeference the generated 3D model.

Incomplete GPS tags are ignored (with latitude and longitude coordinates but without altitude).

GPS altitude reference Sea level and WGS 84 ellipsoid are supported.

Control points should be used whenever you need better-than-GPS georeferencing accuracy, or whenever you want to eliminate long-range geometric distortion caused by numerical error accumulation across the subject's extent. Georeferencing requires a minimum of three control points. Addressing long-range effects demands a higher number of well-distributed control points. The 3D position of control points must be obtained via traditional surveying methods. For each control point, you will have to manually point out a few (2 minimum, 3+ recommended) 2D measurements in photographs through the Smart3DCapture Master graphical user interface or a third-party tool.

See also Control Points.

In addition to GPS tags and control points, ContextCapture can import potentially any other positioning data (e.g. inertial navigation system data) or third-party aerotriangulation results, through position/rotation text file, or through a dedicated XML or Excel format. Once imported, ContextCapture can use this data as is, or slightly adjust it, instead of computing it from scratch. This allows an even higher scalability and robustness.

See also Import blocks.