Hey Guys 'n Gals,
is it possible to match the camera somehow? At this point is is not
correct and I wonder if there is a way to get it working. What I'm
trying to do is, I want to render out a pic with Indigo then render
out the alphachannel with 3dsmax to get rid of the background.
Any thoughts?
Here an example, ... as you can tell, I have to stretch the alphamask
a bit to match the pic, so I guess it is the wrong aspect ratio? ?(
Indigo output: HDRI only; w/o bidirectional support; 32min
Alphachannel rendered with 3dsmax/Vray ...
Comped pic in Photoshop ...
Tried to match the alpha, it's not perfect yet ... but hey it's sunday.
So if someone has an idea how to get it right, it would help a lot! ;o)
take care
psor
[Q] - How to match 3dsmax and Indigo Camera?
Hi psor ..
i think the problem is that you can't move the cameraposition without
moving the environment.. ;(
As far as i found out Indigo seems to built the HDR env... a sphere ...
around the camera .. so your view will aways be the same towards the HDR-
I also had this idea some time ago but it seems the env is fixed around the camposition...
Hope thats what you meant..
If not .. if you mean the lens or FOV.. then ... well nick changed this and
i was kind of frustrated cause i've been try and error for ages until i got the right factor to get it match...before he changed it.
So if you could get me the exact data on how Nick uses the fov...
I could fix it but i don't have time to go for try and error again..
cheers..
u3dreal
i think the problem is that you can't move the cameraposition without
moving the environment.. ;(
As far as i found out Indigo seems to built the HDR env... a sphere ...
around the camera .. so your view will aways be the same towards the HDR-
I also had this idea some time ago but it seems the env is fixed around the camposition...
Hope thats what you meant..
If not .. if you mean the lens or FOV.. then ... well nick changed this and
i was kind of frustrated cause i've been try and error for ages until i got the right factor to get it match...before he changed it.
So if you could get me the exact data on how Nick uses the fov...
I could fix it but i don't have time to go for try and error again..
cheers..
u3dreal
Thanx for the info mate! I meant the FOV ... hopefully Nick
can get the alphachannel into Indigo, this would help a lot.
Because I do a lot of comping lately and since we don't have
some kind of reflection/lighting channel support in Indigo
I have to cut the HDRI out with the mask.
And interesting point you made with the HDRI background,
thanx for that too. ;o)
take care
psor
can get the alphachannel into Indigo, this would help a lot.
Because I do a lot of comping lately and since we don't have
some kind of reflection/lighting channel support in Indigo
I have to cut the HDRI out with the mask.
And interesting point you made with the HDRI background,
thanx for that too. ;o)
take care
psor
Not sure if this is what you're looking for, but...
fov=2*arctan(sensor size/(2*lens_sensor_dist))
The sensor size is the diagonal size, so use the sensor width and aspect ratio.
so I think this will work (haven't tested it)
fov=2*arctan[((sensor_width^2)+(sensor_width/aspect_ratio)^2)^(1/2)/(2*lens_sensor_dist)]
I'm too tired to do the algebra to find what lens_sensor_dist equals given fov. Here's what I get from a CAS - feeling a little lazy .
lens_sensor_dist=(abs((sensor_width)/(aspect_ratio))*sqrt((aspect_ratio)^(2)+1))/(2*tan((fov)/(2)))
really sloppy, sorry ...
Hope that gives you an idea for how to fix it. I haven't tried your exporter, so I'm not sure what's broken. Good luck...
fov=2*arctan(sensor size/(2*lens_sensor_dist))
The sensor size is the diagonal size, so use the sensor width and aspect ratio.
so I think this will work (haven't tested it)
fov=2*arctan[((sensor_width^2)+(sensor_width/aspect_ratio)^2)^(1/2)/(2*lens_sensor_dist)]
I'm too tired to do the algebra to find what lens_sensor_dist equals given fov. Here's what I get from a CAS - feeling a little lazy .
lens_sensor_dist=(abs((sensor_width)/(aspect_ratio))*sqrt((aspect_ratio)^(2)+1))/(2*tan((fov)/(2)))
really sloppy, sorry ...
Hope that gives you an idea for how to fix it. I haven't tried your exporter, so I'm not sure what's broken. Good luck...
HI noby,
thanks alot for all these number... lol
i think it should ba abit easier... as i have local legth..
So when i use :
lens_sensor_dist = (f*d/(f+d)) it seems to fit quite well.. but not exactly..
I came up with a crazy equation ... don't know how even to correct
the focal lengh before lens_sensor_dist calculation.
right now it seems to be very good... but still not perfect with telelenses. Also iz seems that indigo results are somehow moved up and right by one pixel...
@ psor .. feel free to email me if you want to betatest it...
#
cheers
u3dreal
thanks alot for all these number... lol
i think it should ba abit easier... as i have local legth..
So when i use :
lens_sensor_dist = (f*d/(f+d)) it seems to fit quite well.. but not exactly..
I came up with a crazy equation ... don't know how even to correct
the focal lengh before lens_sensor_dist calculation.
right now it seems to be very good... but still not perfect with telelenses. Also iz seems that indigo results are somehow moved up and right by one pixel...
@ psor .. feel free to email me if you want to betatest it...
#
cheers
u3dreal
Who is online
Users browsing this forum: No registered users and 20 guests