Hi, I'm loving the SDK but struggling to integrate with my existing OpenGL application.
I want use the SDK model projection view but not the supplied projection view matrix. If I understood a bit more about matrix math this probably wouldn't be difficult, but the path I took was to start by comparing the CameraCalibration getProjectionGL with my glFrustum. Seems that I can use the SDK pose modelview matrix with my projection view if I scale my projection matrix (1, 1, -1) ie reverse z "polarity". What parameters would I have used in glFrustum to achieve that effect? I've played around with znear & zfar +ve, -ve, larger, smaller but cannot get that elusive model to appear on my screen. It works if I use getProjectionGL.
Any advice?
Thanks
:confused:
I haven't had a chance to dive into the math, but two thoughts:
1) Could you just render your content in two steps, binding two different projection matrices during a single frame?
2) Perhaps you could find an offset matrix between your projection matrix and ours, and multiply the QCAR pose for each trackable by that matrix before using it.
- Kim