DrPablo wrote:
Of course there are other factors at play, because there are plenty of apochromatic camera lenses, as well as black and white images that have a sense of depth.
You lost me a little with the reference to apochromatic lenses (since it is the human eye's focus characteristics, not the lens', that causes the effect) but I agree with the lack of effect in B&W (unless Ansel snuck some blue tint in there ).
DrPablo wrote:
. . . it would preferentially attract our attention to things that are closer and "warmer".
I also was thinking if there were evolutionary causes behind it. You can kind of imagine back in cave man days that "blue mean far; red mean I eat it or it eat me".
We need to put this idea to the test a little bit --
Someone show a 3D-looking image that has a blue foreground subject and a red background. It's hard to find! The sky is generally in the background, not the foreground, and people's skin is generally redder than blue.
This is the only one I can find -- a picture my wife took of me in Red Square, but I think DOF and angular lighting contribute chiefly to the (modest) 3D effect, and there's also a little telephoto compression.
I had mentioned earlier that my point of view is that the key factor is relative contrast between subjects, or relative contrast between parts of a single subject. Nobody had commented on it, so I'll explain further: Air (and dust) is between subjects and lenses. It softens the contrast that let us see the fine details. The more far away there is an objet the more soft it appears. Think of the extreme case of distant mountains, or those images from teleobjectives that are usually flat. Our brain are used to estimate distances by this, and perceives subtle changes in relative contrast. If parts of an image are more contrasty relative to other parts, even subtlely, they appear to be nearer. So the key is not having lots of detail everywhere, but selectively instead. It acts in a similar way like selectively defocusing, but in the real world our eyes (our brain) don't see selective focus and perhaps this is the reason beause shallow focus in photographs also gives us the sensation of depth.
DrPablo wrote:
This is a picture of mine taken with a lens that has fairly crappy bokeh, the Canon 50 f/1.8. I think the picture indeed has a 3D effect. But do you really think it comes from the background? I think it comes from the play of light and texture on the guard's face and uniform.
The lighting and high level of detail on all aspects of the guard from the rifle to the tunic to the face is giving him depth but in addition the slightly OOF background seems to place him firmly in the foreground of the frame. I don't think you would have got the same 3D effect if you had, for example, used a long telephoto and turned the background into cream cheese or done the opposite and brought the background building into sharp focus.
The human eye might not be an apochromat and certainly doesn't focus on blue very well if at all. It is the PhotoShop(TM) in the brain taking care of these. Red text on blue background or vice versa is unintelligible. Not sure if this contributes to the 3D effect but we are talking about visual perception anyway...
I think that the degree of any photo's 3-D effect is determined by how REAL the photo appears to our brain.
The question is what makes a photograph seem real to our brain?
Personally, I think the effect is dependent on how the photographic system handles the interaction of light with the elements in the picture, and how well this approximates how our brain perceives the world through our eyes.
Personally, I think every part of the photographic system plays a role - including lenses, processing - and - the sensor design of the camera.
I have always felt that Fuji P&S's, some Oly P&S's, and, often, Nikon DSLR's show the 3-D effect more than my Canon CMOS sensor. Nikon P&S's, on the other hand, have always seemed to deliver flat pix, to my eyes, at least. Why
Shooting from the hip, I would venture that is has something to do white balance, and the sensitivity and smoothness of the system to the contrasts of light from highlights through mids to blacks.
Since B&W photos do indeed show this 3-D effec, the quality of these lighting contrasts likely have to do with the luminosity aspects, not chrominance.
As other folks have suggested, it may have to do with edges and pixil level contrasts. Interestingly, the Fuji CCD chips have a different pixil geometry than other CCD chips, with a supposedly better accuracy for line edges at 45 degree angles than other chips.
Yet, CMOS-chipped Canon cams can and do show the 3-D effect. Do they require more extraordinary natural lighting then other systems to produce equal effects? Can a simple lens substitution override sensor design?
I wish I knew more about micro- and macro-contrasts and how the human body perceives them. And how various sensors and lenses transmit them, for that is where I think the answer to the 3-D question may well be found.
DrPablo wrote:
We need to put this idea to the test a little bit --
Someone show a 3D-looking image that has a blue foreground subject and a red background. It's hard to find! The sky is generally in the background, not the foreground, and people's skin is generally redder than blue.
This is the only one I can find -- a picture my wife took of me in Red Square, but I think DOF and angular lighting contribute chiefly to the (modest) 3D effect, and there's also a little telephoto compression.
Gingerbaker wrote:
Personally, I think every part of the photographic system plays a role - including lenses, processing - and - the sensor design of the camera.
Don't forget film. Timothy O'Sullivan was making 3D-looking landscapes using orthochromatic films 120 years ago.
This, by the way, is an albumen print. You can still make albumen prints by mixing some silver nitrate and glacial acetic acid with egg whites, then sensitizing with potassium dichromate. All you do after that is contact print under a UV light source. Those chemicals are safe and easy to get. If I were a good enough cook to beat egg whites I might try it myself.
Here's my favorite print of his, made in 1871. The white sky and dark rocks are because of the orthochromatic emulsions available at the time, which were basically sensitive only to blue.
Brent, wow, just about poked my eyes out with that one . . .
and reminded me this is not a new question. Two characteristics in paintings that create a 3d effect are perspective and lighting. Two different approaches as illustrated by Brunelleschi (perspective) and Rembrandt (lighting). DrPablo's blue shirt photo, as well as the guard all have strong side lighting (and a little bit of 'rembrandt' triangle on the shaded eye). The harbor photo shows strong perspective (vanishing point). Background haze (atmospheric or aerial perspective) was also picked up in the renaissance, and is attributed to DaVinci.
Adding selective focus (or selective OOF) is an additional method to 'fool' the eye.
example of aerial perspective (bridge and background mountains lose detail). Not sure if it translates into a websized jpg (medium format printed at 16x20) http://kevin2i.zenfolio.com/img/p49064641-4.jpg
This, by the way, is an albumen print. You can still make albumen prints by mixing some silver nitrate and glacial acetic acid with egg whites, then sensitizing with potassium dichromate. All you do after that is contact print under a UV light source. Those chemicals are safe and easy to get. If I were a good enough cook to beat egg whites I might try it myself.
Here's my favorite print of his, made in 1871. The white sky and dark rocks are because of the orthochromatic emulsions available at the time, which were basically sensitive only to blue.
Brent, nice shot. I still think textures and hues have a lot to do with the 3D look on shots that have all the frame in focus. Some lenses (especially with digital) aren't very good at capturing the texture and hue detail.
pdmphoto wrote:
Brent, nice shot. I still think textures and hues have a lot to do with the 3D look on shots that have all the frame in focus. Some lenses (especially with digital) aren't very good at capturing the texture and hue detail.
I have to say that the cz21 is the best I've used so far on digital.
Alf Beharie wrote:
I dont think colour has any bearing on 3D ness:
It does when it does, it doesn't when it doesn't.
If we've discovered nothing else in this thread, it's that there isn't one unifying variable that all "3D" looking images share. Showing one (relatively) monochrome image that looks 3D doesn't preclude other images from looking 3D by virtue of their colors.
Hi again, I am very sorry I missed some of the discussion. Great to have so many different opinions here.
I made some observations without the use of the camera. In fact closing one eye you may discover scenes which have more 3D and others which apear to be flatter. Turning an object in your hand you may see that 3D look changes with the linghting. I dont think any technique can help when your set up and light is flat, the image may still have other qualities of course.
I' d agree with Paul - Boston that even a cheap lens can give you 3D, if you use your camera wisely. Some scenes offer the 3D look and the photographer needs to catch it IMO
That said I think that there might be a benefit in using MF lenses because we can not shoot so carelessly as with a 18-200 AF Zoom pointing at all directions and shooting whatever moves, but need to make up the lack of autoeverything by skill
Edited by Andi Dietrich on May 04, 2007 at 02:52 PM GMT
pere marti wrote:
I had mentioned earlier that my point of view is that the key factor is relative contrast between subjects, or relative contrast between parts of a single subject. Nobody had commented on it, so I'll explain further: Air (and dust) is between subjects and lenses. It softens the contrast that let us see the fine details. The more far away there is an objet the more soft it appears. Think of the extreme case of distant mountains, or those images from teleobjectives that are usually flat. Our brain are used to estimate distances by this, and perceives subtle changes in relative contrast. If parts of an image are more contrasty relative to other parts, even subtlely, they appear to be nearer. So the key is not having lots of detail everywhere, but selectively instead. It acts in a similar way like selectively defocusing, but in the real world our eyes (our brain) don't see selective focus and perhaps this is the reason beause shallow focus in photographs also gives us the sensation of depth.
).
