Photo matching helper. Camera 2d space constraint with FOV

[aio]

Hello people,

I'm looking for a way to ease photo mapping and image based modeling using ICE.
I found on rray.de an addon called Point_cam_cns but i'm afraid i can't make it work.

Ideally i would like to have something like this inside softimage : http://www.pano4d.com/

Do you if something close to that exist?

For the time being a simple two point constraint will suffice. Basically the user will stick two nulls to the image of a cube's edge.
The two points would stay in place in the camera viewport but as the FOV is adjusted the camera position would change.
This would help determine the picture FOV.

But since i suck at trigonometry and at ICE I don't know where to begin. Any clue or lead?

Thank you !

[dwigfor]

Not sure if I ever uploaded my camera frustum compound... In case I didn't, take a look at Todd Akita's ICE math primer PDF for help building your own. Not sure where on here; it's definitely on Mark's blog.

off the top of my head, tan(angle) equals opposite / adjacent.

I use this on Backpack's eyes and mouth (2d planes), allowing me to adjust z position away from 3d character (preventing penetration), but maintaining scale relative to viewport.

[nunoalex]

Apparently thats one of the new features of Max 2014, pitty no one with the knowledge ever tried to do this using ICE kinematics.