I hope you will soon make the decision to develop further on this software.
I have a few ideas.
I have seen setups, for established systems on the marked, where you calibrate a stereo pair. Imagine a bracket/beam with a projecter in the middle and a DSLR on each side. Would that be a possibility to improve scans…
Furthermore make use of GPU acceleration.
And the obvious, implement control for more camera models if possible.
I think that the demand for a total solution is not the idea - Us 3D-workers can fusion and assemble our meshes in other software. All we need are the good scans.