)
To: visfxsup visfx...@gmail.com,Nuke-Users Mailing List
nuke-users@support.thefoundry.co.uk
Subject: Re: [Nuke-users] Still Image camera matching...?
Nope. It will figure out the lens distortion from lines you draw - or a
grid you shot - or it will attempt to figure out the distortion
richb...@mac.com
Date: 12/09/2013 9:50 PM (GMT-05:00)
To: visfxsup visfx...@gmail.com,Nuke-Users Mailing List
nuke-users@support.thefoundry.co.uk
Subject: Re: [Nuke-users] Still Image camera matching...?
Nope. It will figure out the lens distortion from lines you draw - or a grid
you
Julik,
Yes, I had that in the back of my mind from when I played around with a demo
copy of SynthEyes, a while ago. I’ve had a need for this kind of “reverse
engineering of a camera on several occasions. It would be really helpful to
have this feature in Nuke…
RIch
On Dec 8, 2013, at 1:12
(GMT-05:00)
To: Nuke user discussion nuke-users@support.thefoundry.co.uk
Subject: Re: [Nuke-users] Still Image camera matching...?
I know this doesn’t answer the question “properly”, but I use Syntheyes and
it’s lens lines feature.
Would be cool if Nuke had this.
On 06 Dec 2013
nuke-users@support.thefoundry.co.uk
Subject: Re: [Nuke-users] Still Image camera matching...?
Nope. It will figure out the lens distortion from lines you draw - or a grid
you shot - or it will attempt to figure out the distortion by analyzing a
sequence (it does a camera track first
@support.thefoundry.co.uk
Subject: Re: [Nuke-users] Still Image camera matching...?
Nope. It will figure out the lens distortion from lines you draw - or a grid
you shot - or it will attempt to figure out the distortion by analyzing a
sequence (it does a camera track first) but it needs
I know this doesn’t answer the question “properly”, but I use Syntheyes and
it’s lens lines feature.
Would be cool if Nuke had this.
On 06 Dec 2013, at 18:12, Richard Bobo richb...@mac.com wrote:
Anyone have any good methods for camera matching a single still image in
Nuke…?
--
Julik
discussion nuke-users@support.thefoundry.co.uk
Subject: Re: [Nuke-users] Still Image camera matching...?
I know this doesn’t answer the question “properly”, but I use Syntheyes and
it’s lens lines feature.
Would be cool if Nuke had this.
On 06 Dec 2013, at 18:12, Richard Bobo richb...@mac.com
Hi all,
Anyone have any good methods for camera matching a single still image in Nuke…?
Thanks,
Rich
Rich Bobo
Senior VFX Compositor
Armstrong-White
http://armstrong-white.com/
Email: richb...@mac.com
Mobile: (248) 840-2665
Web: http://richbobo.com/
Man does not live by a turkey in every
First thing I'd do is check if there's any EXIF data in the image from the
camera that shot it. It might give you an idea of the camera make, focal
length, etc. Which will also help you figure out the size of the
filmback. If there's a common object in the shot like a pen or street sign
you can
I sometimes use the ideas in this tutorials:
https://www.evernote.com/shard/s7/sh/70dcb995-c062-4b11-a9de-3a238159ef55/4f2ff05a1d359877f9aea494a05905fe
Ron Ganbar
email: ron...@gmail.com
tel: +44 (0)7968 007 309 [UK]
+972 (0)54 255 9765 [Israel]
url: http://ronganbar.wordpress.com/
On
Im not sure of an automated way.
But can
-Draw perspective guide lines for the image
-drop a cube through a scanline renderer at the image screen resolution
and with correct camera back.
-then can try various different lens until u get a match?
Pw
On 6 Dec 2013 17:24, Ron Ganbar
On Dec 6, 2013, at 12:20 PM, John Mangia j...@johnmangia.com wrote:
First thing I'd do is check if there's any EXIF data in the image from the
camera that shot it. It might give you an idea of the camera make, focal
length, etc. Which will also help you figure out the size of the
Yeah, as soon as you've found out the field of view you can usually match it
quite well by hand. And as the previous poster said, knowledge of the
dimensions of a few objects will help a lot to get the right scale.
If you've surveyed the 3D positions of a few points on set you could of course
Elias,
Unfortunately, the image was shot a while ago, in a remote location…
Thanks,
Rich
On Dec 6, 2013, at 1:24 PM, Elias Ericsson Rydberg
elias.ericsson.rydb...@gmail.com wrote:
Yeah, as soon as you've found out the field of view you can usually match it
quite well by hand. And as the
Well, here I am thinking people live in the best of worlds. I agree with
previous posters, estimated positions of point will have to do. Eliminate as
many variabels as you can. Knowing, or estimating, the FoV is key!
Cheers,
Elias Ericsson Rydberg
6 dec 2013 kl. 19:36 skrev Richard Bobo
Any parallel lines? Find a vanishing point(s). Draw a cube from this and add a
3d cube. Adjust length and rotation of the camera until the cube fits the
lines. Should give you your focal length.
Howard
On 6 Dec 2013, at 18:58, Elias Ericsson Rydberg
elias.ericsson.rydb...@gmail.com wrote:
About that Fxguide article: Anyone know off-hand why the focal length is
derived from a half circle (arc) from the two vanishing points, and the
center of the frame? Why is that the focal length? Everything else is
intuitive to me but that.
Thanks,
Small brain
On Fri, Dec 6, 2013 at 3:12 PM,
Never mind, I got it. Just had to rub my brain cells together a little
longer than usual.
R
On Fri, Dec 6, 2013 at 3:35 PM, Ryan O'Phelan designer...@gmail.com wrote:
About that Fxguide article: Anyone know off-hand why the focal length is
derived from a half circle (arc) from the two
19 matches
Mail list logo
