To solve the problem one needs to transform 2D images of 3D structures from different angles/perspectives into a solid model. This is known as Structure from Motion analysis.
VisualSFM is a software package which in combination with CSVS might be a way to go. On the project website for Visual SFM the author Changchang Wu explains:
VisualSFM is a GUI application for 3D reconstruction using structure
from motion (SFM). The reconstruction system integrates several of my
previous projects: SIFT on GPU(SiftGPU), Multicore Bundle Adjustment,
and Towards Linear-time Incremental Structure from Motion. In
particular, both the GUI SfM system and SiftGPU started as course
projects when I was at UNC. VisualSFM runs very fast by exploiting
multicore parallelism in feature detection, feature matching, and
A youtube movie from www.flightriot.com gives a practical example of potential endpoint products. I have not tried this myself, but will hopefully be able to refine this answer later.
I have also tried using pix4D. It is an overall good tool providing exceptional ease of use for inexperienced to more experienced users. A youtube video introduces the software in a good way.
Dronemapper is an online service where you can upload imagery and have it processed.