This is just standard multi-camera view tiling as has been done in OpenGL for years.
In a two-camera configuration, for example, you usually want the left camera to show you the left part of the view, and the right camera to show you the right part of the view. You do this by multiplying a translation into the projection matrix. If you didn't multiply this translation into the projection matrix, then all cameras would render the same eye space, and you'd get the same image from all cameras. If you tried to do this by modifying the modelview matrix instead of the projection matrix, the images wouldn't tile seamlessly. -Paul > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf > Of santosh > Sent: Tuesday, October 30, 2007 2:42 PM > > Hi Guys > > I am not getting the concept of translate transformation on > projection matrix. When we add the slave camera in viewer we > have to give projection offset. In osgcamera example the > projection offset is given translate transformation matrix. > what it exactly does. > is it translating camera position or rotating the camera? > > Thanks > Santosh > _______________________________________________ > osg-users mailing list > [email protected] > http://lists.openscenegraph.org/listinfo.cgi/osg-users-opensce negraph.org _______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
