Photogrammetry 3D Object Capture Guide
Defining Objectives
What are my user requirements? I want to take photos of a sculpture. I know the maximum limit for a sand sculpting competition is 20x20 feet and while as I’d like to get as much detail as possible, I’m executing for the novelty.
What are my limits? Preference - free and/or open-source software, Hardware - GTX 1080ti sc2 GPU, (Storage N/A as I have many free Terabytes), Phone - iPhone 16 pro
Software Options
Best of Apps, - Modelar (a lot of export options), DepthEye (Export .ply files only), Hedges (special: ability to use front facing lidar), RTAB-Map (only exports in OBJ)
Computer Software - Mesh Room, Reality Capture, Blender, MeshLab
Limited number of captures or freemium - KIRI Engine (150), Polycam (150), RealityScan (300), 3DF Zephyr (50), Scaniverse
Workflows
The two workflows I can either do through the deduction of my options are:
a) Capture raw images and process them through Mesh Room or Reality Capture
b) Capture images through the Reality Capture mobile application for on device processing
If I was in a hurry, I would do option B but since I’m try to get a decent scan I’m going to go with option A.
Guidance for picture taking
How do I get the best scan? “The camera is a factor, but your technique is more important and most cameras will yield acceptable results. The “holy trinity” of photo settings for sharp pictures - camera speed faster than 1/30 of a second to avoid motion blur, ISO of less than 400 to avoid camera noise, aperture set to F8 or higher (smaller aperture) for decent depth of field (but too high will result in distortion). You can bend these rules, for instance you can take slower photos with a tripod, but it may affect your results.”