|
|
|
|
San Diego
179 posts
|
Rather than add to or take over another thread on this forum on this subject, I thought to add this one with some examples.
Just completed a project for a client web site, which included anaglyph renders by me of four of the images. The original drawings were done by Mark Luebker for UCAN (Utilities Commission Action Network) to be posted on their web site as part of a "faux comix" book. I was commissioned to create true 3D images for four(4) of the images to be viewed with red-blue anaglyph glasses. Since many of your don't have these, I am only including one picture here, you can see the rest at www.ucan.orghttp://www.ucan.org/newsletters/the_first_online_comic_book_in_super_magno_x_ray_visionsm.
I am including an overview so you can see the 3D space created with a series of panels in a "shadow box" still set, then shot with two cameras. The output is then composited in Photoshop to create the anaglyph.
I am also including a "stereo pair" just for fun, for those who don't mind going cross-eyed. I used to draw these in art school free hand, but for some people it's headache time. View at your own risk...
The client was thrilled, and their server crashed the first day with all of the viewers trying to see them. They had mailed out the funky glasses in a newsletter that I had also done for them with several 3D images, one of which I will include below.
All done in modo. I loved being able to put in an image mask and see it in the construction immediately. Found some funny glitches along the way. Take a look, let me know.
Paul
An eye for an eye, and the whole world goes blind... Gandhi/MLKing
Greater consciousness is worth the effort. Be aware....
|
|
11/4/2007 - 9:22 PM
|
|
|
Germany
1331 posts
|
Indeed i don't have such glasses, tough i frequently wish i did. The squinting version works fine though, very cool.
|
|
11/5/2007 - 2:35 AM
|
|
|
Cambridgeshire, UK.
5192 posts
|
Absolutely! very cool indeed. Love to know how you did these. I'm guessing the cross eyed one is simply two renders from a pair of cameras aimed at the same point in space? I've no idea how you'd do the anaglyph renders in Modo though!
OS10.6.2 Dual 2.66GHz Dual-Core Intel Xeon Mac Pro, 5gb RAM, GeForce 8800GT
|
|
11/5/2007 - 2:48 AM
|
|
|
328 posts
|
very cool
|
|
11/5/2007 - 4:39 AM
|
|
|
california
76 posts
|
sweet
|
|
11/5/2007 - 4:43 AM
|
|
|
San Diego
179 posts
|
I set up three(3) cameras in modo. The central one for construction and reference, then two separate ones on either side to correspond to the left/right eye renders. Getting them to all focus on the same spot proved to be annoying as there is not, as of yet, a easy way of placing a "null" in the scene and having everybody "target" or "focus" on it, so that part I did by hand. I tried a brief animation, but this aspect makes it a wretched slog, forcing me to move the assets to another program for this aspect.
The work flow started in Photoshop with the original drawings, which I then sliced into component layers, filling in as needed to create the hidden aspects. Each layer was exported as a flattened tif, along with a separate alpha file. Back in modo, flat planes are created, then the material is assigned (diffuse) and the alpha is placed above it and set to "Transparent Amount." What is wonderful about doing this in modo is that I can see the masking update immediately and tweak it as needed. I did have some "weirdness" factor as I angled some planes back into space as my OpenGL would not allow the masked section to show correctly, but it rendered fine, and would usually show up correctly if I saved, exited, then re-entered the same file???
Once a render had been done of all three cameras, I went back into Photoshop for the compositing. For the cross-eyed render, I just placed the left eye render on the right side, and the right eye render on the left side. If you want to do a "wall eyed" version then keep them on the "correct side" but when you cross, it works to have them looking at the correct information.
For the anaglyph version, a different methodology was used. Two of them used color, so here is the drill:
Open the "central" image as the base. Then open the left eye version, SelectAll/Copy, and then paste this into the RED channel only of the base image. Open the right eye version, SelectAll/Copy and paste it into the BLUE channel of the base image. Keep the GREEN channel as is and the color will be somewhat okay, given the severe limits of anaglyph images. For best effect select the RED and BLUE channels individually and move them left or right as needed while viewing the picture with the glasses to get the focus optimal. When setting up the focus point in the scene, remember to place it "intentfully" as the distortion will be generated relative to where the focus point is place. I put it on the characters face, so the feet get more "whacked out." This is not a problem with the "stereo-cross" versions which I love as they are so crisp. I have included another one below for those who can view such things without strain. Enjoy.
Paul
An eye for an eye, and the whole world goes blind... Gandhi/MLKing
Greater consciousness is worth the effort. Be aware....
|
|
11/5/2007 - 5:33 PM
|
|
|
Antwerp, Belgium (EU)
329 posts
|
if you focus a camera on a certain spot, then copy the camera (which is focussed at exactly the same spot) and move its pivot to that focus-spot.. then rotate the second camera using pivot as its action center, you can have'm in different positions focussed on the same spot.
|
|
11/6/2007 - 12:12 AM
|
|
|
Cambridgeshire, UK.
5192 posts
|
Thanks for the explanation of your workflow Paul! I must have a go at this!
OS10.6.2 Dual 2.66GHz Dual-Core Intel Xeon Mac Pro, 5gb RAM, GeForce 8800GT
|
|
11/6/2007 - 12:32 AM
|
|
|
San Diego
179 posts
|
Lunchbox, while this does work in modo, I did exactly this at one point, it is a kludgey work around!!! and heaven help you if you then want to animate the scene, or change the focus of just three cameras, it is a workflow nightmare. It is so simple in other programs, just create a "null" place it in the scene where ever, then target anything at it, cameras, lights, etc. This way you can fly the camera around while moving your object without losing focus. Way simple, I can't wait for modo to include this fundamental element. For the most part, I remain deeply impressed with modo and it's development, but find it oddly informative to find the spots where it stumbles as it tries to grow up. It has been instructive to be this close to a new software as it develops, growing pains and all.
Paul
An eye for an eye, and the whole world goes blind... Gandhi/MLKing
Greater consciousness is worth the effort. Be aware....
|
|
11/6/2007 - 8:41 AM
|
|
|
1977 posts
|
Good stuff. Its funny, I have a pair of anaglyph glasses sitting right on my desk, I saw your post, picked them up and bam, there it was. good stuff.
Yazan
|
|
11/6/2007 - 10:03 AM
|
|
|
San Diego, CA USA
2149 posts
|
Fun project Paul! Maybe bring some glasses and give us a quick demo at our next sdmodo meeting? ;-)
-Jeff
san diego modo user group: www.sdmodo.org
product design+development: www.jeffrutan.com
LEGO fun: www.aerospacebricker.com
business fun: www.timeitnow.com
|
|
11/7/2007 - 2:14 PM
|
|
|
|
|
|
|
|
|
|
|
|
