A quick stereo cam test

[Micha_cg]

Hallo,

Corey here my last quick stereo test. It can be seen in stereo if a mirror faced to the right side is placed on the left side of the white bar. Now your nose must touch the edge of the mirror. If you look now to the left image with both eyes, the left eye see the image direct and the right image see the mirrored image. Your brain should see a 3d image.

Workflow: I have rendered two image from different locations (like the human eyes). Than I have mirrored the right image and placed both in one image.

More infos can be find per google search "mirror 3d image".

For a Vray 3D cam we would need two parameters - eye distance + parallel view / cross view. Parallel means, the camera is shifted and cross view means the camera is turned around the focal point.

[Micha_cg]

A quick stereo cam test

... here an other example. Now the mirror must not touch the monitor. Usage: Place the edge from a mirror between the eyes on the nose - place your head approx. 30 cm in front of the monitor and look at the left image. Now try to combine both images.



The mirror view method is not the only method to see stereo images. The basic for any method are two images like the view of the human eyes. I think, if in Vray could implemented a special stereo camera, than the could render an image like in my examples in one pass.

Here some links about the mirror view method:
http://www.crystalcanyons.net/abouts/3dc7_MirrorL.htm
http://www.crystalcanyons.net/abouts...alMonitor.shtm
http://www.crystalcanyons.net/abouts...rComplete.shtm

http://www.ascscientific.com/stereos.html

general stereo view methods:
http://www.stereo3dgallery.net/3dhelpM.shtm

www.stereoscopy.com

[Micha_cg]

Re: A quick stereo cam test

...

For a Vray 3D cam we would need two parameters - eye distance + parallel view / cross view. Parallel means, the camera is shifted and cross view means the camera is turned around the focal point.
...

In a stereo manual (only german at www.herbig-3d.de) I have read that a parallel shift of the camera is a must. So a Vray stereo camera would need one parameter only - eye distance.

[corey]

A quick stereo cam test

It seems to me that this doesn't need a new stereo camera. The images on the website appear to be exactly the same but just horizontally flipped.

I will try these processes you link to myself as soon as I have a chance.

Best regards,

Corey

[Micha_cg]

A quick stereo cam test

It seems to me that this doesn't need a new stereo camera. The images on the website appear to be exactly the same but just horizontally flipped.

Exactly the same images would not produce a real 3D image, maybe it look like. For example the influence of the eye distance is important for the stereo effect. A higher distance as the eye distance give a stronger stereo effect, but in the worst case a room look like a dollhouse. I want to say, the eye distance is the basic for the stereo effect. How it will be used is a question of the users taste.

An other story: interesting for me was/is, that we can produce a fake stereo effect on a TV set too. Precondition - the TV show moved objects like a flying football or an animal go through a forest. Take a sunglass and place one glas in front of one eye. Now look with both eyes on the TV screen - one eye through the glass and the other direct. Funny effect or?

[Jean-Paul Paloux_cg]

Re: A quick stereo cam test

Hi, any news to produce stereoscopic images with Vray for Rhino ?
except a pan of the camera + post process in photoshop for the red/cyan setup ?
Thank you.
any link for tutorial of this technique ? Thank you,
jp

[Micha_cg]

Re: A quick stereo cam test

VfR news ... wait some months, than it could be a better time.

stereoscopic images ... it depends on the user, what will collected here.

One new idea - the DevalVR viewer can be used for colored anaglyph views for object movies. In the simpliest case two images of a turntable.

[Jean-Paul Paloux_cg]

Re: A quick stereo cam test

Thank you Micha,
but what I am looking for is the technique to use to have the same result as in your example from my 3d model. how to place or render, what are the steps in photoshop if we need a post process. Does it affect the output color of the scene viewed behind the glasses red and blue. and how to minimize the impact of this colorisation.
Thank you.

[Micha_cg]

Re: A quick stereo cam test

I don't know so much about stereo imaging. Only I know cross and parallel view are possible, the distance between the camera determinate the depth effect (if the effect is to strong, than the scene looks like a doll house, try to use a natural eye distance). How to create red/blue images? I don't know, I used the function of DevalVR only.

[frip51]

Re: A quick stereo cam test

Just a quick guess:
creating redish/blueish images could be achieved with tweaking the white balance of the physical camera.

[Micha_cg]

Re: A quick stereo cam test

Here some infos and there is a photoshop tutorial:

http://nzphoto.tripod.com/sterea/anaglyphs.htm#intro

[Jean-Paul Paloux_cg]

Re: A quick stereo cam test

Yep, Thank you !
I will use that also: http://nzphoto.tripod.com/sterea/anaglyph_make.html
and I will use place camera and target (same target point) and two camera separated by the distance between my two eyes (3.5") with a 50mm lense.
JP