3D sterio stacked scorpion

Images taken in a controlled environment or with a posed subject. All subject types.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

tpe
Posts: 478
Joined: Sun Aug 26, 2007 4:07 am
Location: Copenhagen Denmark

3D sterio stacked scorpion

Post by tpe »

Stacked in CombineZP 6 deep at f8 1/60 105mm sig macro. Then my new little pet (looked after by the GF, thanks for putting up with it) was rotated slightly different angle. I guess this is not the correct way as the lighting then changed a bit? It is my first time at sterio and I am quite thrilled it worked, also that my model stayed so still for the whole procedure (It had to be repeated a couple of times because the flash was giving some odd exposures). It is really quite a friendly scorpion, i have been prodding it about into the right position with my fingers and it hasn't even tried to sting yet. Landscape is not a good format for 3D, but i didnt find that out till just now, portrait would be much better as it will fit on the screen better, so i am afraid a lot of detail has gone in the resizing. There is a slighly larger, but still screen res friendly one here.

Image

Hopefully he or she will stay still for another session in landscape format, as there is not even a fly around at the moment it is so cold here. I am guessing it is Euscorpius italicus (found in tuscany at about 300m) but it is very small at about 2cm with its tail unwound.

Thanks for looking and if there are any problems with the picture showing up please press F5

tim

Ken Ramos
Posts: 7208
Joined: Thu Jul 27, 2006 2:12 pm
Location: lat=35.4005&lon=-81.9841

Post by Ken Ramos »

Tim wrote:
It is really quite a friendly scorpion, i have been prodding it about into the right position with my fingers and it hasn't even tried to sting yet.
Just give it time Tim, just give it time. The minute it thinks your not looking, then... :shock: Really a great stereo subject though. When folks first started posting them I had a time seeing what everyone else was but now I finally go the hang of it and you know what? It gives me headaches. :lol:

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Tim, welcome to the wonderful world of stereo stacking! It looks like you're off to a good start.

All your comments are dead on.

Rotating the subject under fixed lights does introduce some disparities that have nothing to do with depth and are therefore confusing. I generally describe these as producing a "shimmering" effect. Other viewers may see them differently, depending on eye dominance. In this pair, there are quite a few places that shimmer, especially on the critter's left side legs. With shiny subjects, you can get these problems even by moving just the subject. That situation is improved by more diffuse lighting.

To see these effects, often it helps to layer the left & right images in Photoshop, then click between them quickly and repeatedly using the little "eye" button in the Layers control panel. This lets you see differences in glare and shadows as movement instead of stereo disparity.

There is one peculiar thing in this pair that I can't explain. On the left-hand image, there is a darkened bar extending on the 11-to-5-o'clock axis across the stinger and rear leg. It looks like a shadow, but I can't see anything in the image that would be casting a shadow. Do you have any ideas?

One last observation -- in the right-hand image, the foreground is lighter and the background is darker than on the left. I suppose that reflects some change in the lighting?

--Rik

tpe
Posts: 478
Joined: Sun Aug 26, 2007 4:07 am
Location: Copenhagen Denmark

Post by tpe »

Ken Ramos wrote: Just give it time Tim, just give it time. The minute it thinks your not looking, then... :shock: Really a great stereo subject though. When folks first started posting them I had a time seeing what everyone else was but now I finally go the hang of it and you know what? It gives me headaches. :lol:
You are not kidding it is beginning to give me headaches too, I couldnt get the GF to see it and she got a headache anyway ;).

Thanks for all the info again Rik, I don't know how you have the patience sometimes, a wealth of info. I had a go at another one where i managed to put the lights and the subject on a table that rotated, then moved the whole thing together and it seems to have sorted the shimering, i didnt realise that a simple background is a real help and the scorpion didnt stay as still this time i think as i seem to get some kind of shake on the stack and the background was too chaotic and i got loads of nasty stacking artifacts so will have to do it again i think. If i get anything out of the stacks i will be sure to post. It takes a fair few goes each time as the scorpion tends to move in one or other of the stacks about 70% of the time.

The odd thing that you cant explain is I think due to the age of the flash system i am using, a very old and basic ring flash and diffuser. It has trouble keeping an even output and i think the lines are due to different exposures for different layers in the stack. I had a go at manually adjusting the exposure bfore stacking in PS but it wasnt perfect. I have just tried it with a different flash and it isnt great either, I will have another go with some studio flashes and umbrellas (I thought it a bit overkill at first) and see if that helps. So yes i think that the stripes and forground are both due to the variation in the flash output. :). I had never noticed the inconsistancies in the field but using the centon (m20?) ring flash is perhaps not recomended. I will post as and when i get a better shot :).

thanks

tim

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Tim, you're very welcome for the information. I get some value from thinking hard enough to write, so it's one of those win-win situations.

A couple of things I just noticed...

1. This topic really should have been posted in the Technical and Studio Photography forum. I'll move it over there.

2. I notice you spoke of adjusting the brightness "in PS" (Photoshop?) but you're stacking in CombineZP. The "Align and Balance Used Frames" function in CombineZP is supposed to take care of brightness in addition to doing geometric alignment. Be sure to use that. If you don't, then you're liable to get many nasty artifacts around edges. Also, be aware that the pyramid maximum contrast algorithm is not very sensitive to brightness changes anyway.

Stacking live specimens is always challenging. Even when they look like they're sitting dead quiet, they're liable to shift position just a little bit when you're not paying attention. The moth that I did for Figure 1 of this article gave every indication of holding completely still for hours. But when I looked again at that stack, just a couple of days ago, I noticed that in fact the critter did a couple of small pushups a couple of times, right in the middle of the stack.

About the background, you're right that plain is good. There's an issue that arises in closeup & macro stereo that just doesn't happen in street & landscape stereo. The issue is that in street and landscape stereo, background disparity is strictly limited to a fairly small value. Whatever disparity there is for detail at infinity, that's the most there will be anywhere in the background. But in closeup and macro stereo, the background disparity is essentially unlimited. It's very easy to get so much disparity that background features refuse to lock up at all. Deep stacks make the problem worse by making the background sharper. If you reduce the separation to what's needed for the background, then you also reduce the stereo effect for the subject. A better solution is to make sure that there's just no detail in the far background, which is why plain is good.

--Rik

tpe
Posts: 478
Joined: Sun Aug 26, 2007 4:07 am
Location: Copenhagen Denmark

Post by tpe »

Yes Rik i noticed that too, but only after i had already posted it, funny i just dont think of it when i am using a macro lens as opposed to a microscope or other fixed stacking setup and a live specimin, i thought i was out in the field not in the studio :).

First i tried align and balance used frames but i got some odd artifacts and then used the strict alignment, as the new help file (nice all these updates we are getting :)) mentioned it might be solved that way. On one of the two stackes it worked very well but on the other it gave some islands. I get the feeling that it doesn't balance the exposure either? For the next stacks i used a more modern flash and it has given better results but the background has caused some more problems with islands and is too iregular to work well, but i have used the align and balance used frames and it has worked pretty well except a bit of halowing around the front claws. Please excuse the darkened border (in ps) as the background was just horribly distracting, and i think this way it "locks up" better. I think it has worked better this time and the lighting seems much more pleasing to me. It is from a single overhead flash throuhg a circular diffuser mounted on a tripod, you know one of those collapsable white semi transparent bits of material and gives very nice soft shadows and plenty of light. Going to use a plain background next time though :).

1200 pixel wide one here.

Image


Cheers

tim

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Tim, this is much improved! There is hardly any "shimmer" in this pair. The only places I see are these:

1. under the head and right pincer (as seen) -- image on the left side has a shadow that is much lighter, causing detail to be present that is very dark or absent on the right side.
2. tips of pincers -- right side photo looks unfocused there, like the stack was short by a frame or two.
3. left side, left pincer (as seen), there is a crisp white hair that appears very fuzzy if at all on the right side.

That bit of OOF foreground debris in front of the right pincer is unfortunate. It's hardly noticeable in single views, but pops right out in stereo.

These are all minor defects, but it sounds like you want to hear about whatever I see.

By the way, this busy but OOF background works well for me. The disparity is still small enough that it fuses up with no problem, and it adds some more depth because, well, it obviously is a lot farther back.

Show more, please! :D

--Rik

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

Rik
Do you have any tricks for getting fusion in non crosseyed stereo.

I can use cross eyed instantaneously.
On the scorpion pictures I put my day book in front of my nose to prevent me from crossing and could not fuse the images.

I had trouble with stereograms at first too I tried to attempt to relax my eyes but it didn't seem to help.

Gene

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Gene & Tim, I am embarrassed I did not mention...

The first pair of pictures in this topic is arranged for "parallel" viewing, with the left eye image on the left and the right on the right. To see it properly on most monitors, most people will need a stereo viewer.

I can fuse it OK on my laptop, where each image is only a little over 8 cm wide, but that's only because years of practice have convinced my eyes that it's OK to go walleyed (divergent). Even I can't manage enough divergence to get it on my larger monitors.

However, the second set of pictures is arranged to be viewed cross-eyed. (Gene, have you tried those that way?) Cross-eyed will work for much larger images, especially with experienced viewers.

Gene, I'm not aware of any tricks for fusing parallel stereo. If the image size is more than eye separation, it's almost impossible because people have spent a lifetime learning not to diverge.

Tim, when you're preparing stereo pairs for unaided viewing, it's far better to arrange them cross-eyed. If you set them up for parallel, you have to be sure that each image ends up not more than eye-separation apart on the viewer's monitor. That's a maximum of around 6 cm, 5 cm is safer, only around 180 pixels so it's painfully short of resolution. See http://www.photomacrography.net/forum/v ... php?t=3908 for an example showing both ways.

One last point, Tim. I notice that you're hosting images on your own server, in a directory named "_temp". I'm concerned that these images will disappear at some point in the future. If that happens, then the value of our discussion in this topic goes way down because the examples are lost. For stuff like this, I'd really prefer that the images get pushed over to the photomacrography.net server so they get archived along with all the words. Ship me a PM if that's a problem for some reason. Thanks!

--Rik

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

I did wonder about the second image because I am easily able to view cross eyed images and the relief was not backwards as it is in the first if you view it cross eyed.

So I will stop looking at the second one and try again on the first.
I can make my eyes do a few tricks but I haven't worked on wall eyed ness.

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

g4lab wrote:I can make my eyes do a few tricks but I haven't worked on wall eyed ness.
There are tradeoffs. It's nice to be able to handle slightly over-size parallel pairs and even some vertical parallax. But it is a bit awkward when I go in for an eye exam. Every time the doctor sets up for a convergence or tracking test, I have to explain my somewhat unusual training and ask what she's looking for, how she's going to use the result, and what a "normal" person would see. Otherwise she's likely to get a rather strange impression of what my glasses need to be!

--Rik

P_T
Posts: 461
Joined: Sat Jul 19, 2008 1:13 am
Location: Sydney, Australia

Post by P_T »

You prodded it with your FINGER?!? Are you mad?!? :shock:

The second stereogram is much easier to look at and the angle is better as well. :D

Post Reply Previous topicNext topic