Tony[B] wrote:

>> - setting up camera by hand is very hard and long task, I'll try to
>> write some engine for extracting camera params from photo as part of
>> POVarch...
> 
> 
> I would suggest you do this by having the user draw some lines around the
> edges of the other building(s) and then creating a box from this so the user
> knows which rotations to use for the CG buildings.

In my mind, it was something, which allows you to click into image and 
write 3D coordinates of clicked point and after you click n points, it 
creates POV camera definition for you.

Post a reply to this message