z-Depth from video

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

THREAD CLOSED
 
Thread Tools Search this Thread Display Modes
Old 03 March 2013   #1
z-Depth from video

Is there a way to create a z-depth channel from a panning video shot?
 
Old 03 March 2013   #2
Best way with a panning shot is to use a time offset of a frame or two to get left and right eye. Of course depends on the shot.
 
Old 03 March 2013   #3
huh interesting method
how would you go about getting that into a z-depth clip that you can use in AE for DOF?
 
Old 03 March 2013   #4
Nuke X has this feature ... the node is call Depth Generator .. you need a solve camera then you plug the node into it ... you can give it a shot.
__________________
LDEA Kon Kuatica
 
Old 03 March 2013   #5
It's nearly impossible to create depth from a true nodal pan though, you pretty much have to invent it.
 
Old 03 March 2013   #6
Originally Posted by Rareden: Is there a way to create a z-depth channel from a panning video shot?

It's possible to estimate depth from motion, but only as long as there is parallaxing in the shot (video explaining parallax). Panning is the process of rotating the camera in place, so there is very little parallax. A shot with dollying (moving back and forth along the view) or trucking (moving sideways) can extract a much better depth estimation, since the motion of the objects in the frame is mostly because the camera position is changing, rather than the camera rotation.


Edit: Also, depth estimation does not work so well if there are objects that are moving independently of the camera, like cars or people or what have you. I haven't used the depth estimation tools in Nuke, but I'm sure they have some way of dealing with this. Just beware that it can cause problems.
 
Old 03 March 2013   #7
Originally Posted by CaptainObvious: It's possible to estimate depth from motion, but only as long as there is parallaxing in the shot (video explaining parallax). Panning is the process of rotating the camera in place, so there is very little parallax. A shot with dollying (moving back and forth along the view) or trucking (moving sideways) can extract a much better depth estimation, since the motion of the objects in the frame is mostly because the camera position is changing, rather than the camera rotation.


Edit: Also, depth estimation does not work so well if there are objects that are moving independently of the camera, like cars or people or what have you. I haven't used the depth estimation tools in Nuke, but I'm sure they have some way of dealing with this. Just beware that it can cause problems.


In nuke or any other software you just have to MASK the moving elements
__________________
There are no men like me, only me
 
Old 03 March 2013   #8
Thread automatically closed

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.
 
Thread Closed share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright 2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 11:03 AM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.