Hello!
I’m currently making photogrammetry rig for head scanning. I placed cameras onto vertical bars with ~40 degree horizontal angle (5 cameras onto each bar and 9 bars total). I found that with focal length about 35mm RC cannot make alignment between corresponding horizontal pairs (vertical are ok because of much smaller vertical angle).
By looking at other rigs I see that this problem can be solved in two ways (picture attached). Which way is better for using with RC?
Hi Ivan123
The B example for camera arrangement is much better suited for RC.
Hi Wishgranter
Thank you very much for your answer. I think that I understand why B example is much better at camera alignment stage - because of much more corresponding features. But what is counter-intuitive for me (and that is a main reason of my post) - how example B may be better at reconstruction step? Each of vertical pairs in example A hash much more similarity between images, and I suppose that stereo matching must perform better on A.
Hi Ivan123
It’s better not only for alignment but even for reconstruction.
To understand this you need to dig in DISPARITY calculation to get an idea why it’s better and why are the stereo-pairs technology from the 90´s.
Thank you for pointing me!