Static or dynamic reconstruction?

[albb20]

Hi,
I want to reconstruct an object from my own data. The object was captured by 3 fixed position cameras while falling. Naturally, this is a dynamic scene, i.e. a video. But it would also be possible to handle it as a static scene, where the object is fixed and the cameras are moving around the object. I have only 6 - 12 frames per camera.
With which method would you expect a better reconstruction quality? Can you give me a tip a priori or should i try both?

btw: The images are greyscale, which should not be a problem if i repeat the grey channel twice to get the RGB-shape, right?

Thanks in advance!