jaysdesk 11:01am, 12 August 2008
Caution: This whole subject is too esoteric, read geeky, for the majority of the anaglyph group members. Hardcore anaglyphers (the ana-gly-phisti?) however are cordially invited to comment, advise, and otherwise extend technical or moral support, to this cause.

I just posted the second floating window anaglyph, the first one posted 3 days ago has already been relegated to back pages of the group pool, thanks to the vigor and productivity of the members. I believe the subject matter is important enough to warrant a discussion thread.
The first floating window image: www.flickr.com/photos/fromjaysdesk/2742982149/
If the stereo window is forced to float in front of the monitor screen or the print, and everything else is kept behind this window, then objects can be placed ahead of the screen whether or not these elements intersect the stereo window's borders. This is a big deal. It means objects can pop out of the screen without the need for the gymnastics of keeping them away from the borders of the picture. People can still stand on the ground even when they are in front of the screen. And they can do that without violating the stereo window and pissing off the stereo window purists! Imagine the freedom this brings to composition and framing of 3D photography, where the ground can act naturally and approach the bottom of the picture!
Another advantage of the floating window is, that the world is not seen through an opening that appears to have been cut off the screen, giving the eyes a place to rest and keeping the screen in optical and mental focus.
The stereo window detached from the screen is a phantom object and doesn't capture the eyes, leaving the eyes free to glide back and forth in the picture, thereby enhancing the realism of the scenery, which is the object of 3D photography in the first place.
Sadly I've not been able to overcome the technical difficulties of floating windows. For example I don't fully understand the relationship between a raised window and the additional margins needed on the fringes of the picture. Such understanding is crucial in saving "real estate" of the stereo window and in preventing unnecessary ghosting.
What is needed is straight forward procedures and methodology to allow everyone to experiment with floating windows.
I'm eagerly waiting to hear your thoughts on this subject.
The link to the floating window anaglyph: www.flickr.com/photos/fromjaysdesk/2756442576/
Ian5281 Posted 10 years ago. Edited by Ian5281 (member) 10 years ago
This is not new, and some folks have really made the most of the idea. I would love to use the concept if it were EASIER! I have done a few experiments with adjusting the stereo window with results that I was happy with... but it took too much time. If someone had some easy-to use templates that could be used in SPM.....then I would be much more interested! I always have an errant tree branch or the back of someones head I would like to slice out of the frame.

Here was my ultra-quick response to a comment from Shahrokh on yet another one of my window-violation laden anaglyphs.... I just hacked the shape of the window, without moving the chips relative to each other at all!


No Window Violation

Here I spent a little more time.

Study in stereo window placement

BUT... the man you want to talk to is Roger Harris... who has done stunning work with the stereo window....
www.flickr.com/photos/roger_harris/

Oh.. I should mention... what you do depends on your goal... if one wishes to eliminate anaglyph ghosting, it is often necessary to put the highly contrasting areas at screen level... and floating windows can be useful for this. Odd shaped stereo windows are great for making the most of the composition, or removing unwanted tree branches. Floating the window out in front for it's own sake.... that is up to you.
I have seen that using a neutral background helps reduce the ghosting that results from strong applications of the technique.
SeanGleeson 10 years ago
All of my "Antique 3-D Cheesecake" series use a floating window. It might not be what you have in mind, because these windows are not simple shapes, but ornately illustrated frames, which even cast a shadow on the monitor-plane. But they do cause some of the effects you're talking about. They make the picture seem all the more dimensional. That woman's elbow comes right up to the monitor-plane, but with the floating window it still looks set back.

Sisters: Antique 3-D Cheesecake No. 4
jaysdesk Posted 10 years ago. Edited by jaysdesk (member) 10 years ago
Thank you Ian, Sean, for your attention.
I'll get into the technicalities of floating the window, after its usefulness is established, and more responses come in.
Someone out there may have already come up with practicable procedures and helpful ideas.
By the way, floating window was handed to me by Photoshop through sheer dumb luck , while I was testing an idea on Sean's club_penguin anaglyph. I let the photoshop do the alignment as part of my work flow.
This time photoshop chose to overlap parts of the image instead of the frames of the chips, floating the window in the process. www.flickr.com/photos/fromjaysdesk/2742210997/
Significantly, I ran the same photoshop action on the photoshop aligned pair as I do for all of my anaglyphs, noticing the raised window after wards. This incident suggests that automating the process may not be that difficult.
Anyway, I can not even claim the method as my idea!
SeanGleeson 10 years ago
Gosh, your new penguin pic escaped my notice 5 days ago, I will add that to the rest on the page.

I just noticed, all of the anaglyph examples in the 3D Gugle gallery have floating frames. Maybe that is something his program does automatically by default? (I don't have that software, and don't plan to try it until there is a Mac version.)
jaysdesk Posted 10 years ago. Edited by jaysdesk (member) 9 years ago
Sean; thanks for bringing the Gugle Gallery to my attention. Although I'm not going to do any serious conversion anytime soon, I learned some valuable lessons, just by examining these anaglyphs.
The first lesson is that it pays to place the main subject on, or very close to the plane of the screen. Most of the anaglyphs in these galleries are so clean, that by looking at the thumbnails, it's hard to tell that those are red-cyan anaglyphs. We tend to push everything back, and place the front-most object or the violating object on the screen plane, to prevent the violation, missing the advantage of clean and ghost free anaglyphs (at least in the main subject area).
In all these anaglyphs, the placement of the main object and the bogus frame, seems to be his deliberate choice, rather than any requirement. Obviously he does lift the window, but only where he places the faux frame, maybe as a diversionary tactic, probably to enhance the depth perception, I doubt that window violation has much to do with his choices.
By the way I find his cubist painting conversions quite interesting.
Dan (aka firrs) 10 years ago
Hey guys. I've always been in favour of placing screen depth in the middle ground of the image, at the point of main interest, or location of maximum ghosting...because I think it helps people who are not used to seeing anaglyphs and because I find it very difficult to converge images with large separations. I started out that way, but eventually moved over to placing the image behind the screen because purists complained about it.

So I wonder if this is an acceptable compromise. Here are some images with strong 3D processed so they float:

Bubbles

Bubbles

Pangea Circus

To make them is quite simple, but involves some extra steps. I used StereoPhoto Maker. Line up the images as normal with the front at screen depth. Save the pair of images and (in another application) add a border that is wider than the image to both (mid grey is good for avoiding ghosting). Then make your anaglyph placing the depth where you wish.
neil.nathanson 10 years ago
I've been wanting to try one of these for a while. Here is my approach, make the window a window box and float it...

Speed Racer 3D Stereo Window Box
jaysdesk 10 years ago
Dan, Neil; you guys are way ahead of me, when it comes to 3D window manipulations, however I'm going to ask you to take a look at my ideas commited to a primitive blog spot, and offer advice, as to how to make the process, easy enough, so it can be adopted by anyone who wants to give it a try. fromjaysdesk.blogspot.com/2008_08_01_archive.html

Neil; I love what you did with that window box. It puts everything that I have done with stereo windows to shame.
neil.nathanson 10 years ago
Jay, thanks, lots of really cool and inspiring work from everybody in the group. Let's all keep playng with the idea of the window, there is more to do I am sure.
Carlsson 3D 10 years ago
I agree with what Neil just pointed out, to achieve the understanding that Jay mentioned. Wonderful guys !!! A hole lot of reading though... :)

It is no secret that I´m no fan of frames in general, but I am a huge fan of doing in order to understand. Exactly like you guys are doing here.

Around here, or at least in my circles, computergenerated frames are the very essense of bad taste. Manmade ones are not, when made to be an integrated part of the image ofcourse. I am also aware that Danes, from Denmark, often tend to be in opposition to the rest of the world, so I won´t hold it against you if you feel different about it.

My own personal reasons are.. that not only does one add something "unnatural" to the image, it very much becomes not only a part of it, but most often a more significant overall factor.. And, when applying a frame in front of the stereowindow you need a uniform frame to frame it. Now you got two frames. If the uniform frame isn´t the same as the interface in which it is being presented, you got three frames. The image is now reduced to playing 4th violin in its own symphony. Sean´s sample works really well on the white Flickr interface, try looking in slideshow mode to see the three frames when the interface is black.

Dan/firrs, I like your new correct use of the stereowindow ! Do fool around but please don´t ever go back.
SeanGleeson Posted 10 years ago. Edited by SeanGleeson (member) 10 years ago
Let me enumerate the three ways I agree with Louis (and I'm not even Danish). The garishly shaded frames in the Gugle gallery strike me as especially tacky and distracting, which is a shame because the images inside are very good. If you're going to do a frame, that is not the way to do it.

It is also true that a floating frame forces a choice of background color, so if it does not match the page background, it effectively creates yet another frame, as Louis said. I think it should only be done on a page with the same background as the one behind your frame.

And finally, it is true that the frame becomes as much a subject of the image as the things in the picture. If this is a deliberate choice, I have no problem with it. In Neil's example, and Ian's, and my own, we obviously meant to call attention to the frame itself, and make it the star, or at least co-star, of the show. If you are trying to use it without calling attention to it, I don't expect it will work out that way.
neil.nathanson Posted 10 years ago. Edited by neil.nathanson (member) 10 years ago
Oh one more thing, As Ian pointed out above go look at Roger Harris' frames, he has done quite a lot of them, and this is one of my favorites, It reminds me of a snow globe.



And here is a whole set of fun frame experiments!
www.flickr.com/photos/roger_harris/sets/72157604432436014/
jaysdesk 10 years ago
Thank you all for your participation in this discussion. Roger Harris' work is indeed inspiring, something that every ambitious stereo photographer and artist should study carefully. However I was hoping that the group's effort be focused on finding the least common denominator, the absolutely simplest way of displacing the stereo window, to prevent violation and to gain some flexibility in placing the elements with respect to the plane of the screen, or the printed paper.
Dan's approach is elegant in its simplicity. I wonder if trimming the gray surroundings, harms anything, saving some window real estate in the process.
My unfortunate choice of word in describing, Gugle guy's frames was not meant to reflect a value judgment, and none should have been inferred.
I prefer to leave the elevated stereo window's borders alone, yet a few pixels wide stroke, may make the anaglyph easier to view, something to be tested.
Carlsson 3D Posted 10 years ago. Edited by Carlsson 3D (member) 10 years ago
Jay, the best advice I feel I can give you is tell you to be aware of your stereowindow when shooting, at all times. If damage has allready been done, try bending the violating area back into space with a dephtmap.

What you are refering to in Gug´s gallery would be the size of his spaceimage. It is very small, the images are not very deep. Dan/firrs will love it I´m sure. There is a reason for it ofcourse, and it is not to avoid ghost created by high contrast deep in space. When I played with Gug some years ago I learned a thing or two about his method. He was converting in one image only. When you do that there are certain bounderies to how much distortion the image can take before it starts to get ugly. When I convert I divide the total distortion, or displacement if you like, into two images. That way my spaceimage-size can get much bigger. Same thing with DanD3Dman and Wagman, their good old layered cardboard approach can provide you with a much larger and more natural spaceimage-size.
Dan (aka firrs) Posted 10 years ago. Edited by Dan (aka firrs) (member) 10 years ago
I agree with 'computer generated' frames on pictures being generally tasteless. For that reason I prefer to think of my approach as proving a 'mounting' for the picture :-)

The mounting is there to enhance the viewing process - I put a white border around my cross-views for the same reason.

You can't predict what colour background your image will be viewed against so you can't trim exactly around the image. I tried a less wide rectangle in the examples above, but I don't think it worked so well - seems you need some width on the mounting to keep the raised window effect from breaking down (aesthetically at least) at the edges.
neil.nathanson Posted 10 years ago. Edited by neil.nathanson (member) 10 years ago
Sorry I missed your original intent, so OK how to make floating a window repeatable and easy for all.

My first thought is it is not that easy now. I do Anaglyphs in StereoPhotoMaker it is 1 click easy.

So how about if we define "Floating a Window" and request the developers of StereoPhotoMaker to add it to the software. I bet it would be easy for them to do. I think it simply means allowing the left and right edges to overlap in the easy adjust window instead of clippping the extra stuff away when you move the uppermost slider left and right.

That would make it 1 click easy for everyone.
Did that make sense or am I missing something?
jaysdesk Posted 10 years ago. Edited by jaysdesk (member) 10 years ago
Neil; I presume your question is addressed to me. Unfortunately I'm handicapped by complete unfamiliarity with SPM. Maybe Ian can pitch in and answer a couple of questions about the capabilities of SPM, especially the panoramic plug-in that does the precise alignments.
1. Can the pano application be coaxed to do what the auto-alignment script does in photomerge, namely, expand the size of the individual chips, to the size of the "panorama"?
2. And if so does it allow, the manual placement of the chips? (photoshop doesn't allow the latter. You have to accept its placement.)
increasing the size of the individual chips to a canvas size, large enough to accommodate both chips at their shifted position is the unfamiliar extra step needed to lift the stereo window.
This , I believe is the obstacle to be surmounted, appending, each chip with a white or transparent wing at the opposite end.
It took me a while to gather my thoughts and come up with a meaningful description of a floated window. I was tempted to say, it's a matter of shifting the frame of the stereo window away from the plane of the screen....., but how do you consistently frame a 3- dimensional object? So I decided to settle on this definition:
A -. A floated window is a NORMALIZED stereo window whose frame is shifted away from the screen or printed paper.
B -. A normalized window is a stereo window corrected of window violations, that acts as a virtual frame through which we view 3-d objects, which objects may in turn protrude out of this frame.
sheesh..
Dan (aka firrs) 10 years ago
It simply needs to do what I described above:

1) Align as normal
2) Crop the stereopair for 'correct' window placement with the front object at screen depth
3) Allow the user to then adjust the screen depth without cropping occuring
4) You need to be able to set a colour for the background - and I think add a wider border for it to work nicely.

Should be fairly easy for the developers to implement in SPM I think - it already does way more complicated things than that.
neil.nathanson 10 years ago
Yes Dan's technique is simple and we are both saying the same thing, but he has actualy shown the result in his example above.

One question to Dan, would your approach work with any color border instead of a gray one. Or could the "border" color be the background you predict it wil be on. So if you plan for a black background it could be switched in the SPM program.

So next we should invite David Sykes to this thread. He is one of the developers of SDM and SPM. Anything else we neeed to state before we do that? I can reach out to him if you guys think we are ready.
Dan (aka firrs) 10 years ago
The choice of (sensible) background colour* doesn't matter Neil - and yes you could set it to your page background colour and have an invisible mounting. I chose mid grey because it ghosted less than white.

* Obviously a colour like pure red would be a silly choice...

Would be great to have that added to SPM...already a brilliant piece of software. Can't think of anything else needed.
Mike Bittner 10 years ago
The tool built into SPM for anaglyph window issues like the ones shown here is a fuzzy border. To eliminate ghosting in anaglyphs, some subscribe to the school that you align the image to the main subject of focus and blur the edges of the frame. I have been using this technique on problem images for many years by using feathering in PSP; after I created the anaglyph, I copied and pasted it as a new image with this border effect. I have found the technique useful when making holiday cards and sending them to people who are just casual stereo viewers. See:

stereo.jpn.org/eng/stphmkr/fuzzy/fuzzy.html
Dan (aka firrs) 10 years ago
Yeah - A fuzzy border helps with retinal rivalry for objects at the edges behind the screen depth as well... but is not the prettiest solution.
jaysdesk 10 years ago
I'm sure David Sykes would instantly understand what we're talking about, as soon as he gets exposed to this bloated thread. He may be even able to show us a workaround for shifting the stereo window itself, without the need to change a single line of code.
His expertise would benefit all of us. So, Neil, by all means please do invite him to this discussion thread, and group. It would be edifying to see his anaglyphs and other works too.
neil.nathanson 10 years ago
Cool, I never noticed fuzzy windows, I'll have to try a few.

I sent email to David, let's see his input.
neil.nathanson 10 years ago
I heard from David, He said he would be glad to participate and see what he can do. But he is in the middle of a persona photo project photographing emerging caterpillars from eggs and needs some time. So we can pick this part of the topic up again when he and mother nature are ready.
microfunguy 10 years ago
Hello folks , you have not been forgotten !
I am still working on the butterfly project, building equipment, solving problems, etc.
It will soon be finished but the creatures will not wait.
If you are interested, here is a stereo timelapse clip (2.8 MB) :
www.zshare.net/video/183585868ab3e9b1/
I need to start the sequence earlier and take synched photos at faster intervals (one second).
Dan (aka firrs) 10 years ago
Looking good so far David.
microfunguy 10 years ago
Still filming caterpillars but I have played with this a bit.
Assuming that we have loaded a pair of images into SPM that are already adjusted to the window ... display them as an anaglyph and press left arrow key until red/cyan images overlap on the the object you want at window level.
'Position alignment' at bottom of screen tells you the 'x' shift.
Assume it is +12 pixels.

Let us assume that a new SPM command 'Float Window' is clicked.
It will add a border of say 50 pixels to the outside edges of the unshifted source images and a border of 50-12 to the inside edges.

Does that do what you want ?

(I used PSP to add the unequal borders to the images)


David
jaysdesk Posted 10 years ago. Edited by jaysdesk (member) 10 years ago
Thank you David for helping us out here. I'm away from my home computer and can not test anything for the next few hours, yet I wanted to respond to your post in a timely manner.
Both chips need to be expanded by a minimum of the 12 assumed pixels, the left chipon the right side, and the right chip on the left side.
I personally question the value of addition of a random border, as it further shrinks the area of the sereo window available to picture content. But this is an issue that we should have settled, before we came to the Patron Saint of the Anaglyphisti - you!
Dan (aka firrs) 10 years ago
Yes thanks David. I'm trying to get my head around a description in words, but that I'm pretty sure that's what we want.

jaysdesk the additional border width is a must in my opinion. I've done a few of these now - try cropping off the "extra" bits youself and see what effect that has. The retinal rivalry that results is really not comfortable.

Fireworks by Dan (aka firrs)
microfunguy 10 years ago
Well, try it out Dan.
Set the zero deviation in SPM and note the x-shift.
Add suitable margins in PSP (or whatever) and then load into SPM and view as anaglyph.
jaysdesk 10 years ago
Dan, I may be over protective of window real estate, but once a color is chosen to append the chips, it can not be simply trimmed off- the fringes will end up with wrong tint that doesn't get properly filtered out.
Here's a collection of all of my floating window anaglyphs: www.flickr.com/photos/fromjaysdesk/sets/72157606693669302/
If you and others find this approach uncomportable to view, I'll respect that opinion.
Dan (aka firrs) 10 years ago
OK David, I've tried it step by step.

The x-shift is negative (because you are pulling the image towards you). Otherwise it's as you say - but probably best to change the maths so you can't accidentally get cropping (shift greater than the mounting).

So assume it is -12 pixels and you want an extra 50 pixels 'mounting' either side.

You need to add:

A rectangle of 50+12 pixels on the left of the left image (outside) and 50 pixels on the right of the left image (inside).

A rectangle of 50 pixels on the left of the left image (inside) and 50+12 pixels on the right of the left image (outside).

To be most useful the colour of the background should be user selectable as well as the width of the mounting and border top and bottom.

jaysdesk - I don't fully understand what you are saying. In your examples the background is white and so on a un-modified flickr page looks ok...but try looking at the same image on a black background (e.g a flickr slideshow). You can't predict what colour the picture will be displayed against and if its not the same as the colour you've chosen for your background you get uncomfortable retinal rivalry. Follow this link and see for yourself what I'm talking about...

www.flickr.com/photos/fromjaysdesk/sets/72157606693669302...

I choose mid grey for my background as pure white and pure black can cause a little ghosting.
microfunguy 10 years ago
Dan you have said exactly the same thing in a different way !
The 'x' displacement at the bottom of the screen was positive when I brought the subject through the window.
I will check it again later.
I saved the test images with unequal mid-grey borders as BMP's and in PSP selected the grey borders and replaced with a 'parchment' pattern.
Worked fine.
For now, if you wanted to narrow the borders I guess you would crop the same amount from left and right sides of test images.
jaysdesk 10 years ago
I'm sorry I didn't make it clear that white is my uncontroversial choice of background color. I occasionally print my anaglyphs and give it out to people. I also do not consider flickr slideshow to be the ideal mode of presentation of my or others' pictures. It either hides the title and caption of pictures or clutters the pictures with those.
Dan (aka firrs) Posted 10 years ago. Edited by Dan (aka firrs) (member) 10 years ago
jaysdesk - Yes of course I can see why you might want to use white and have no 'extra bits'...but I was answering your question as to why you might also want to have 'extra bits'... :-)

David - when I originally made these manually through PSP I don't need to bother working out how much extra to add. I just added a generous amount equally to both left and right and then aligned and cropped the image in SPM. The extra bit of workflow is therefore not a major pain ...but it would indeed be nice to be able to do it from within SPM.

Edit -Yes I was basically agreeing with you David, but there was a subtle difference with what I said . If the shift is 60 pixels and the border is 50 then the maths as I wrote it doesn't result in a -10 pixel canvas increase...but yeah - you described what we want.
microfunguy 10 years ago
Dan, the 'x' value is negative as you say.
This is quick and easy to do in PSP but I will add it to list of possible SPM additions.
However, Masuji's work takes him away from home for very long periods and there are more urgent 'fixes' required in SMM so do not hold your breath !
Dan (aka firrs) 10 years ago
I do also use SMM and I would agree that is the priority. The various 'everyday' workarounds for producing video are a lot more time consuming.
microfunguy 10 years ago
Going briefly off-topic .. I will be making my first movie but have not yet decided on a workflow that includes SMM's auto-correction.
You seem to need massive amount of temporary storage.
I think some of my clips represent the limit of magnification for stereo movies because of the exaggerated stereo-depth caused by the diverging angle of the microscope lenses and the fact the creatures do not like the massive light intensity (heat is filtered-out).
Even though lens spacing is a massive 1/5th of subject distance, I find this 8.7MB cross-eyed clip can be comfortably viewed :

www.zshare.net/video/18758423f4c71241/

(eggs are 2mm high, caterpillar is 3mm long)

The 'Cabbage White' is still laying eggs on nasturtiums and other plants.
Dan (aka firrs) 10 years ago
I presume you mean that your microscope lenses are converging on the subject meaning you can't have much distance into the background.

I read about an imax camera rig that can go from zero stereobase upwards in a single take. Apparently the cameras shoot into what is essentially a one way mirror (one above and one behind).

I don't find that example awfully good 3D. Maybe it's because the video is only small and loses the details.

(Have you tried uploading to vimeo?)

Video in general requires a lot of storage - and when making anaglyphs for flickr I tend to go through intermediate stages of uncompressed avi so the colour channels don't get contaminated. Fortunately I've got nearly 3TB :-]
microfunguy 10 years ago
At that magnification you have to attach external aperture stops to get some (still very small) depth-of-field.
That reduces resolution (due to diffraction), there is no way around the problem for movies.
For still images you can make an image-stack with a single wide-aperture lens and combine into an extended DOF image.
Dan (aka firrs) 10 years ago
I meant distance into the background without the stereo blowing up to massive separation. The small DOF might actually work in your favour in that respect. Its a tricky one though - have you thought about getting a piece of half silvered glass so you can set the cameras parallel? I'm intrigued to know if it would work.
microfunguy 10 years ago
neil.nathanson Posted 10 years ago. Edited by neil.nathanson (member) 10 years ago
Hey Guys, I was just checking the latest version of SPM it now has a floating window feature!

I know this looks silly, just showing what is possible. I applied the window effect 4 times just for experimenting.

This was prompted by Jay's Desk's suggestion and David Sykes helped get the request to Masuji and he added it.

Try it out.
stereo.jpn.org/eng/stphmkr/index.html


Float that window in SPM
jaysdesk 10 years ago
Please click on the above picture for more comments and issues.
jaysdesk Posted 10 years ago. Edited by jaysdesk (member) 10 years ago
Well folks, here it is, a powerful implementation of Floating Windows in version 3.25 of SPM stereo.jpn.org/eng/stphmkr/index.html
You'll find this feature under edit menu of SPM.
I wish to express my gratitude to Masuji Suto and David Sykes, who added this feature to an already powerful application, that many of us use to produce our anaglyphs. This implementation of floating windows, includes all the features that we asked for and then some.
I would also like to thank everyone who participated in this discussion thread, especially Dan (aka firrs) and Neil Nathanson , whose contributions directly resulted in inclusion of floating window, with all of its desirable features, in the latest version of SPM .
Now we all have a powerful tool to fight window violations and ghosting, and to add excitement to our anaglyphs. Happy window lifting everyone!
Dan (aka firrs) 10 years ago
I've just tried it and it does the business. Good work...Thanks to Masuiji and David.
Stereotron Posted 10 years ago. Edited by Stereotron (member) 10 years ago
Without having read the whole topic, I agree that an arbitrary shaped frame
is always an artistic approach. I don't know how this can be generated by the computer. It would depend on the developers cleverness how the
outcome will look like.
I take the time by shaping paths so they cut away the 'bad' disparity
and hope it is fun to follow the 3D contour.
Groups Beta