Very Sexy 3 color DCG

[Johnfp]

Dinesh, thank you for the clarification. It seems I take words litteraly and not figuratively. That is, pseudo really means false color. And I truly get that if you looked at my model after I painted it, it looks nothing like the hologram - color wise. But if I were trying to make a hologram of a model such that I could sell the model, then my hologram does justice to the original model.

But with all that aside, I think we all have to do certain things to get our holograms to come out looking realistic. Isn't that the ultimate goal when doing a known object? We dont want a person's (or model's) skin to be green anymore then we want the grass in a scene to be gold. So to me, no matter what technique you use to get familliar objects to their know color it is a good thing in holography and one that the 3-laser full color producers try to strive for.

[Dinesh]

It seems I take words litteraly and not figuratively.

John, I don't think it's the difference between "literally" or "figuratively", it's really a case of nomenclature. I remember one of the first books I read on holography that said that there are numerous names for any kind of hologram. For a simple reflection, for example, you could call it:
single- or double-beam Denisyuk volume dielectric phase reflection hologram (and I can probably come up with a few more if I thought about it a bit more!).

In order to keep life relatively simple, we make up metaphors, or 'handles', so that when we use these metaphors, everyone (hopefully) knows what we mean. So, 'pseudocolour' and 'real colour' as referred to in holography are both really false colours insofar as nothing has a 'real' colour'. A white object in sunlight seen under moonlight looks pale blue, an object seen in yellow light looks different when seen in white light (I remember my physics teacher in high school recommending that when we bought clothes, go out into the sun to see what it 'really' looks like because some stores exaggerate the yellowness of their light). But a 'colour' hologram - and they're all coloured, but some are more coloured than others - is always the same colour because it's caused by diffraction. If you see the colour at all it's because the hologram is receiving the colours that are already "within" it. Look at your hologram under green light and her skin will look as yellow as her hair. Look at it under red light and you won't see anything at all, or hardly anything. There's almost no red fringes in your hologram.

The point I was trying to make is that words like 'pseudocolour' and 'real' colour are matters of technique, not vocabulary. By using words like 'pseudocolour' and 'real' colour, you're telling the world: "This is how I made this hologram".

I'm still thinking of submitting a paper to ISDH on colour maps for diffractive processes. If you're coming, we can talk of this some more.

[Johnfp]

I will do my best to go and at this time I am about 80% sure I am going.

The point I was trying to make is that words like 'pseudocolour' and 'real' colour are matters of technique, not vocabulary. By using words like 'pseudocolour' and 'real' colour, you're telling the world: "This is how I made this hologram".

One correction. If you use those words that are matters of technique, you are not telling the WORLD, you are telling others that know of the technique. Very subjective. The real world will go out and look up the definitions of the words and take them literally. So someone off the street that is not a holographer will go look up Pseudo and see it means false. Then they will look at the hologram and wonder what is false about it, not knowing the technique. That was my point exactly. But again, I do get it because I am posting on a forum of not the whole world but surely a group of people that know the techniques and to that the words have to be chosen in context of the understanding of the audience and not the literal meaning. I do get it and will try to be more accurate in my descriptions in the future.

[Dinesh]

One correction. If you use those words that are matters of technique, you are not telling the WORLD, you are telling others that know of the technique. Very subjective.

True. When I said "world", I meant the holographic world. I think the general public doesn't really care how it's made, they look at the end result and if that looks like it's got colour, then it's got colour. Asking whether or not it's pseudocolour or real colour is a bit like asking whether you use Olive oil or Mazola (vegetabe oil) when you fry.

In holography, I think that the perception of technique is important in that we (the holographers) judge the work not just by it's image quality, but the techniques that were used and the difficulty of those techniques. The fact that you found the right paints, determined the right amount of exposure and determined the right amount of processing shows you thought long and hard about technique. As Jeff Weil points out, I bet you learned a great deal from all of that! So, for me in particular, the fact that's it's a doll (sexy or otherwise) is not as important as the steps you took as a result of analysis, creativity and determination that went into the technique to create the image. In this vein, just calling it a pseudocolour or just calling it a full colour does the image a disservice, because you discovered a new technique that blends bot those traditional techniques. The exact number of lasers is really not that important. As far as I'm concerned these discussions on the number of lasers for "accurate reproduction" is nuts! You have three receptors in your retina and, no matter how many colours you use, the brain "sees" the colour as a result of data passed to it by three receptors. So, asking how many lasers were used is a bit like asking how many vegetables you put in Minestrone, or how many spices you put in a roast lamb! Oh, by the way, I think I can prove that if you do use 4 or more lasers, you have an infinite number of possible colour maps, and so an infinite number of gamuts. So 4 is better than 3 only because you have a far greater number of choices - an infinity of them, in fact!

[Johnfp]

Jeffery, I could not find the hologram, can you belive it. I looked everywhere. But I did find another one that I took a film of that show paralax of about 160 - 170 degrees. Of course it distorts when that far out to the side. I can e-mail it to you if you wish. It is rated R. She is topless. It is not a 3 color one.

I did find a slightly longer clip of the color hologram with a little more paralax. Just put your player on auto-repeat.

http://www.holograms3d.com/PhotoGal/Mod ... Longer.wmv

And I did find the model, her dang foot was broken. Here are two images side by side of the hologram and the painted object.

ModelColorCompare.jpg (37.7 KiB) Viewed 3567 times

[a_k]

Her hair is Yellow, her top wrap is green, her dress is orange and her skin, well is skin color.

But you seem to have recalled the original colours incorrectly:

Her hair is rather "peroxidblonde", her wrap clearly is blue, the dress is yellowish ocker.

Now the only resemblence colourwise is her skin but she seems to have had too much sun and fallen into an olive colouration bath before the hologram was taken. On the video at certain angles her knee seems to have the correct colour tone. Am i missing something? Is it the video which is considerably off with the colours?

[Johnfp]

Now I'm confused. Are you saying that her wrap is blue in the hologram?

You have to read the technique to understand the color changes. I wanted to do multiple colors in one scene in DCG. I wanted one of the colors to be as close to red as possible. So in recording with the longest wavelength of my Argon laser, that would be 514. Process swelling of 16% gives about 596, pretty close to orange-red. Now to get some green with that same swell of 16% I had to use the shorter of the wavelengths of my Argon Ion laser. 457 swelled 16% gives about 530 which is green.

Now how do I go about getting just a reflection from the blue 457 wavelength. Well I needed a paint that reflects 457 but does not reflect 514. Turns out to be Cobalt Blue. And how do I get reflection from just the 514 but not the 457, that was found to be Cadmium Yellow. So I have green and I have reddish. Finally what paint reflects both the 457 and 514 wavelength which when swollen is 530 and 596 which turns out to be yellow, that is Titanium White.

As you can see choosing the correct paints was essential as the two wavelengths of my laser 457 and 514 are really not that far apart as paint reflections go.

I didnt want to make a hologram with 457 adn 514 and no swell as that would have just given me blue and green and a blue green for both. Blue, green and blue-green are so close that it would have looked like pretty much a single color. Now swellign those wavlengths, gives green, yellow and orange. A lot more color then the originals.

[a_k]

Sorry for the confusion. I thought the picture was the original model but apparently it was taken after the painting. Is there a picture of the model before it was painted?

I've read the tests on your website and appreciate that it takes an accurate control over the process to get predictable colour results.

[Dinesh]

So what you're saying is that you paint your model so as to get "close" to some specific colour, then aim for the colour by application of paint, recording wavelength and processing. In the model, presumably you intended for the wrap to be close to red, ie red shifted. You achieve this by painting the object blue so as to maximise reflection at some wavelength. So, after painting, the wrap is blue. If the model was now exposed with a blue wrap and the wrap came out green after recording and processing you have achieved the aim of getting it "closer to red", insofar as the green of the image is closer to red than the original wrap.

It seems that what you're doing is getting an algorithm for transforming one specific colour into another, rather than reproducing a specific colour on a real model into that same colour in the final hologram. So, to get a green colour, as on the wrap, you start with a blue colour, as in the original wrap, then you go through a specific process involving wavelength, exposure and processing - effectively an algorithm - and so effect a transform that may be specified as :

Paint Blue ->shoot with some lambda -> process to get some specific swelling-> green

This would make the original colour irrelevant, insofar as the original colour is not represented by the painted colour. That is, when you paint, you've lost all data of the original colour. To all intents and purposes, the "new, painted" colour - the colour you're actually shooting - is now blue.


I think that may be the source of the confusion. A colour picture of any kind implies a reproduction of the colours in the original to the colours in the reproduction. So, if I take a picture of a grass-covered lawn, I assume that the photograph of the grass covered lawn looks pretty similar to the real grass covered lawn. However, if I wish to translate the green of the grass into a blue grass (for whatever surreal reason) then I might effect such a change by some choice of dye that translates a green image into a blue one. In this case, I'm also taking a colour picture with my camera, but the camera has imbued in it an additional function of colour translation. However, if were to lend my camera to someone and he/she were to take a picture of his/her lawn, she might complain that the green grass became blue in the final picture. The argument that it's still a colour piccture is, of course, valid, but the miscommunication arises because of the interpretation of the words "colour picture".

[a_k]

If you look at it as a black box:

There is a model with yellow hair, green top wrap, orange dress, some processing takes place, a hologram results with yellow hair, green top, orange dress.

So the models original colours are (partly) reproduced somewhat accurately. The model is not her self anymore afterwards, but you've got the hologram, so who cares