Not at my machine atm but off top of my head have a look at these commands
set current camera x
x=object in screen(x)
used together I think the detection should work
set object mask x,x
this will only show object to specified camera until the mask is changed
however they won`t orientate your planes to face the correct camera, instead I think you will have to work it out for each plane between each frame after detection, then update the orientation of each plane followed by setting the object mask.
Alternative to object mask would be to use hide/show/exclude object on/off, object/object limbs between viewport syncs. As there is no limb in screen() command you would have to use a seperate object that is positioned and orientated then removed from render after detection for each limb if you went the limb route. Another pair of commands that you may find some use for are object visible and limb visible commands which only relate to show/hide commands - you can have an object thats excluded but not hidden and the object/limb visible command will still tell you if its hidden or not
Also idk what code you are using for viewports but I did ask lee for an example some years ago and I can acces that email from this machine so here is the example so you can check if its the way yours is set up similarly;
set display mode 1024,768,32,0
set window off
sync on
sync rate 60
make object sphere 1,1
`viewports
backdrop on 0
color backdrop 0,rgb(255,0,0)
make camera 1
backdrop on 1
color backdrop 1,rgb(0,0,255)
make camera 2
backdrop on 2
color backdrop 2,rgb(0,255,0)
make camera 3
backdrop on 3
color backdrop 3,rgb(0,255,255)
set camera view 0,0,0,screen width()/2,screen height()/2
set camera view 1,screen width()/2,0,screen width(),screen height()/2
set camera view 2,0,screen height()/2,screen width()/2,screen height()
set camera view 3,screen width()/2,screen height()/2,screen width(),screen height()
position camera 1,0,5,0
point camera 1,0,0,0
position camera 2,-5,0,0
point camera 2,0,0,0
position camera 3,0,0,-5
point camera 3,0,0,0
do
for t=0 to 3
if t=0
set object wireframe fsysobj+1,1 : set object diffuse fsysobj+1,rgb(255,0,0) : set object specular power fsysobj+1,1000
sync mask %1000 : fastsync
endif
if t=1
set object wireframe fsysobj+1,1 : set object diffuse fsysobj+1,rgb(0,255,0) : set object specular power fsysobj+1,1000
sync mask %0100 : fastsync
endif
if t=2
set object wireframe fsysobj+1,1 : set object diffuse fsysobj+1,rgb(0,0,255) : set object specular power fsysobj+1,1000
sync mask %0010 : fastsync
endif
if t=3
set object wireframe fsysobj+1,0 : set object diffuse fsysobj+1,rgb(255,255,255) : set object specular power fsysobj+1,100
sync mask %0001 : fastsync
endif
next t
sync
loop
There is a way using shaders to orientate the planes to camera but I do not know at this point if something would need passing to the shader to tell it which camera to orientate towards, additionally I have no clue as to how using point object for particles would affect performance versus making the calculations in dbp yourself, but cheap solution might be to take the current camera angle and rotating each particle to the same angle plus 180 degs, but this might not work if you start zrotating the camera
