Photogrammetry – Photographs become 3D

Photogrammetry is not a new invention. However, only in recent years has it been able to establish itself as the perfect 3D scanning method because the texture (colored surface) of an object can only be kept in a scan with photogrammetry.

Due to the technical developments of the last decades, computers for personal use are so powerful now that this compute-intensive method can be widely used. During photogrammetric scans, distinctive points of an object (color, geometry, etc.) are clearly identified from several perspectives. The software then calculates the position of these points in a 3D space. The outcome is a so-called dense cloud. Connecting these points results in a 3D model. Finally, the texture is projected onto this object and everything can be rendered in the end.

With this quick description, you will already realize the possibilities as well as some sources of error:

• Photogrammetry is based on the precondition that several photos capture a point from multiple perspectives (angles). It is therefore necessary that the photos are shot around an object. They should have an overlapping of 60-70% so that the features on one photo can also be recognized on the next one.
• In addition, the photos must be detailed enough to recognize as many features as possible. In order to achieve this, you need good illumination and, if possible, no unessential features in the space which contribute little to the calculation.
• Of course, the object should be constructed in a way that a feature can be recognized from several angles. You will likely reach the limit of possibillities with reflective surfaces, transparent surfaces, pure black or white surfaces, or shadows.

Only an interaction of optimally coordinated processes can minimize these sources of error and crown a calculation with success. This is solely possible with the 3D Apparat.

Our construction is like a fully automated photo studio. With a glass rotary table, objects can be captured fully automatically from all angles in just one go. The consistent studio light allows you to perfectly crop the individual photos as basis for the photogrammetric calculation and to deliver optimal photo quality.

After the software (we use PhotoScan/Metashape) has sorted the photos and calculated a 3D object, you can submit this 3D object for further processing. Countless fields of application are waiting for this data to capture reality and make it virtually usable.

It is our aim that you do not have to worry about anything except replacing an object and scanning the next one at the push of a button.