clang and gcc complained about this, while VS2015 accepted it. It turns out this is the much better way to do it!

Added SOP algorithm to features, added Python bindings section, updated GLM path and added note to clone with --recursive.

Submodule points at release tag 0.9.8.3

I think what happened earlier is by merging the old python-bindings branch first, the pybind "upgrade" was actually a downgrade. But now it should be back to the latest.

* Now using the transparent glm bindings
* Replaced nonlinear camera estimate with the linear algorithm
* Updated RenderingParameters bindings

Adjusted RenderingParameters constructor accordingly.

This is useful to get correct angle estimates when, as in OpenCV, the viewport is upside-down.
It could maybe also be solved by negating the 2nd and 3rd row of the rotation matrix in RenderingParameters, but I couldn't get the projection right. This is probably nicer.

It still needs a more proper cleanup.

RenderingParameters gets more and more chaotic, needs a major overhaul.

Based on the affine camera estimate, and then doing SVD to obtain the closest orthonormal rotation matrix.

This is mainly to accommodate fit-model-ceres, where I've used double for all parameters

At least gcc-5.4 on Ubuntu 16.04 needs it. It shouldn't hurt on older versions.

At least gcc-5.4 on Ubuntu 16.04 needs it. It shouldn't hurt on older versions.

Careful about GLM's notation! A 3x4 matrix is a mat4x3!