More than one approach to 3D and Stereo photography

Stereopsis and Aging

Wright LA, Wormald RPL. Stereopsis and Aging. Eye 1992;6:473-6.

The largest population- age based study yet performed (albeit in 1992 in the United Kingdom) was performed on 417 adults 65 years-of-age or older. In a nutshell here are the results: Twenty-seven percent had normal or better BDP (defined as 55 seconds of arc). Seventy-three percent had reduced acuity of depth perception (BDP) ranging from just below normal to and including the 29% who were totally “stereo-blind”.

I have mentioned this published research report before. However, given the frequency of criticisms from at least one forum monitor, as well as from a few others regarding their personal inabilities to foveate vertically and diagonally displaced disparities as my presumed failure to assure that “only horizontal disparity displacements” should be used in 3D image pairs, I think that further discussion of the topic is warranted.

It seems, therefore, useful to emphasize that such a recommendation concerning horizontal displacements is being made by individuals suffering from age-related visual defects of a variety of sorts that result in diminished abilities to foveate any disparity displacements other than those of the horizontal sort that most of us were able to foveate as infants. Most of us experience a rapid phase of improvement in our BDP (including foveational virtuosity enabling foveation of vertical and horizontal disparity displacements - as well as of horizontal displacements) through our early twenties.

Our BDP continues to slowly improve into our fifth decade. Statistics reveal that sometime in our forties, our matured BDP acuity begins to wane as a consequence of one or more effects of advancing age in about 73% of us. Asymmetric presbyopia, cataracts, ocular neuromotor incoordinations (strabismus, amblyopia- “lazy eye”)and use of “near and far” contact lens pairs are all common causes of decreased BDP acuity. Anaglyphs are often the last resort of some of our older cohort to get the last of their 3D “kicks”.

I do appreciate those individuals who, in spite of the supposedly knowledgeable (albeit erroneous) comments of some forum members, express their individual ability to successfully perceive the intended 3D effect of the 2D-to-3D conversions that I have posted. After all, 27% of Wright and Wormald’s test samples had normal -or better- BDP. It is good to know that I am not alone in having lucked out being blessed with good BDP into my eighties- and, I hope, longer!

I am well aware that the 3D conversions I post are critiqued by fans of traditional two-shot “stereography” for presumed inadequately strongly differentiable foreground depths, continuance of discernible depths extending deep into distant backgrounds, and a lack of “cardboar…”…oh no; sorry, they don’t mention that!

In fairness, I ought mention that most fans of 3D conversions particularly appreciate the smooth continuance of discernible depths far beyond the point of “horizontal displacement extinction” of three dimensions. This is thought “more natural” than leaving all beyond 150 to 200 yards as a flat, 2D painted stage backdrop or museum diorama background that is a consistent feature of ortho stereography.

To which comment they quickly respond that all that the two-view stereography bunch has to do to extend their limit of 3D is to more widely separate their lens axes by a few centimeters, feet, meters, or even miles in the case of aerial stereography! (Sorry…too much “fussing” for most photographers, although I must admit that occasional traditional “two-view” ortho stereographers are careful not to include deep background depths and thereby with some image pairs accomplish a high degree of fine art 3D imagery!

Those favoring 3D conversions are typically traditional photographers who “work the scene”from multiple perspectives, often leaving a generous “surround” to permit compositional virtuosity in creative cropping. It is they who often, in reviewing their archives, are struck by the thought: “Damn, I’d sure like to see THAT in 3D!” And so, of course, they can!

I do find it strange that any illusory perception (which, by the way, any 3D perception is) can be considered more “real” or “natural” than another 3D perception. What is “normal” or “realistic” to some clearly is not so to others. As with any illusion, no two individuals perceive them in exactly the same manner. Vive la difference!

IMO it is ridiculous to suggest that two-view ortho stereography is really the only 3D format welcomed or considered to be “real” 3D photography, simply due to the evidently predominantly older age demographic of this particular forum.

And there, I’ve said my piece. I do hope the result is to make our forum more welcoming to both realms of 3D photographic imagery.

And for those who hate digging out pertinent literature:

Here are a few samples of 3D conversions from 2D originals:

Best regards,

Dave Graham

uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Dave,

I, too, am disappointed by the lack of depth provided by conventional stereo and like landscape's to the horizon if possible, wherein one mountain can be discriminated from another. Over the Festivities, with time on my hands, I decided to try out hand painting depthmaps for several of my own old 2D landscapes and greatly enjoyed myself, discovering much more about 3D on the way. Examples below (all Xeye).

tony brown wrote:

Dave,

I, too, am disappointed by the lack of depth provided by conventional stereo and like landscape's to the horizon if possible, wherein one mountain can be discriminated from another. Over the Festivities, with time on my hands, I decided to try out hand painting depthmaps for several of my own old 2D landscapes and greatly enjoyed myself, discovering much more about 3D on the way. Examples below (all Xeye).

…and….here are the image pairs I mentioned:

uuglypher wrote:

…and….here are the image pairs I mentioned:

Strangely, the old, rusting Glasgow Puffer is one of the most prominent and easily seen in 3D by others.

In the pictures immediately below I see only the top pair well and by Xeye.

In the centre image, it doesn't work for me.

In the bottom image only the top pair work for me and still as Xeye.

The result is that what works for me when making my own images, may well not work for another viewer.

Unfortunately, I am unfamiliar with your terms and abbreviations and work mainly, as an empiricist, by trial and error.

Aged 83 in 5 weeks time, I may well be on totally the wrong track but am thoroughly enjoying my interest in 3D.

Hi, Tony,

You state:

”The result is that what works for me when making my own images, may well not work for another viewer.

Unfortunately, I am unfamiliar with your terms and abbreviations and work mainly, as an empiricist, by trial and error.”

I am in total agreement on all points; there is no sense in disbelieving what our own eyes and mind perceive, no in disbelieving others reporting differing perceptions!

Those terms and abbreviations are simple codes for the several hundred possible disproportionate transformations and reminders to me as to the particular geometric transformation used to accomplish the desired disparities in the right-eye image. My process for deciding on an approach to any particular image is first to decide upon the appropriate  direction of depth recession (DR) to emphasize by promoting detail size recession(DSR) in that direction. There invariably turn out to be - if I may mix a metaphor - “more than one way to skin a 3D conversion”. 
I am also continually aware that the 3D illusion, like all illusions, is not perceived identically by any two viewers! Ernst Gombrich’s book “Art and Illusion” contains many illuminating perspectives on the psycho- physiological aspects of visual “perception”, on of the most trenchant of which (IMO) is that of “guided projection”, also termed “the viewer’s burden”.

With all this in mind, I have prepared another set of two 3D conversions , each arranged for viewing by both parallel and crossed gaze, and am most curious as to your take on them.

And I am wishing a most happy and prosperous New Year to you, my friend!

Dave

Tony,

Having just re-read my last missive I must admit to the necessity to change “…several hundred possible disproportionate transformations.” to “…an unlimited number of transformations, depending upon the fineness of granularity of differences in strength of any particular transformation.”

Ah…there….now I can sleep tonight!

Dave

uuglypher wrote:

Hi, Tony,

You state:

”The result is that what works for me when making my own images, may well not work for another viewer.

Unfortunately, I am unfamiliar with your terms and abbreviations and work mainly, as an empiricist, by trial and error.”

I am in total agreement on all points; there is no sense in disbelieving what our own eyes and mind perceive, no in disbelieving others reporting differing perceptions!

Those terms and abbreviations are simple codes for the several hundred possible disproportionate transformations and reminders to me as to the particular geometric transformation used to accomplish the desired disparities in the right-eye image. My process for deciding on an approach to any particular image is first to decide upon the appropriate direction of depth recession (DR) to emphasize by promoting detail size recession(DSR) in that direction. There invariably turn out to be - if I may mix a metaphor - “more than one way to skin a 3D conversion”.
I am also continually aware that the 3D illusion, like all illusions, is not perceived identically by any two viewers! Ernst Gombrich’s book “Art and Illusion” contains many illuminating perspectives on the psycho- physiological aspects of visual “perception”, on of the most trenchant of which (IMO) is that of “guided projection”, also termed “the viewer’s burden”.

With all this in mind, I have prepared another set of two 3D conversions , each arranged for viewing by both parallel and crossed gaze, and am most curious as to your take on them.

And I am wishing a most happy and prosperous New Year to you, my friend!

Dave

I can see in both cross eyed versions, which is how I am viewing them, some correct stereo effect in both. The depth is in the correct direction.

The parallel versions seem strangely flat, unlike twin lens stereos when viewed by the wrong process i.e. cross eyed viewing of side by side images, which have a conflicting view to the eyes.

However, though I can make an estimate of real depth with twin lens stereo images, with your versions I have to keep renewing my mind's eye view of them and can not assess depth. To me it could be any figure I try to put on it. The depth is truly suggested but not quantified.

Pointless quotation: "Our West window has all the exuberance of Chaucer with none of the concomitant crudities of the period." Alec Guinness: 'Kind Hearts and Coronets'.

Hi, Tony,

Here are two similar, but different transformational treatments of the same original image (graciously contributed by Lawrence Smith) that have generally been well-received by my usual referees ( fully able and competent free-viewers by either technique, but, of course, each has his or her favorite one).

It was α 50-50 wash as to which yielded the “clearly superior 3D illusion”

Care to put these under your X-gaze and register your opinion as to which ( if either) is judged “superior” - and, if possible, why?

Many 5hanks for any input you may offer--

Dave

uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

uuglypher wrote:

Hi, Tony,

Here are two similar, but different transformational treatments of the same original image (graciously contributed by Lawrence Smith) that have generally been well-received by my usual referees ( fully able and competent free-viewers by either technique, but, of course, each has his or her favorite one).

It was α 50-50 wash as to which yielded the “clearly superior 3D illusion”

Care to put these under your X-gaze and register your opinion as to which ( if either) is judged “superior” - and, if possible, why?

Many 5hanks for any input you may offer--

Dave

uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

I viewed them both full size as Side by Sides and frankly didn't get 3D out of either.

The depth I saw was merely the perspective built into the photograph but as flat pictures. Strangely, the lower one seemd more forward to me than the upper one BUT that was the whole photo rather than the tree trunk alone.

Incidentally, I have changed my technique for using Tri-shots and 3rd image joins are not on the surface now. (Xeyed.) These have been re-edited and no longer need the water to join the third image in.

How do you see them? 3D or not 3D, in the words of Hamlet?

They are just narrow angle, twin camera stereos but with long distance 'deep stereo' discrimination input from a Tri-shot.

Thanks for looking and commenting, Tony.

In the  first pair of yours the depths with strongest discernibility are - to my eye- in the middleground (water, curving wall, sw@n) but some, less strongly defined depth intervals extend to the horizon.

In the second pair I see discernible depths as far as the near water’s edge; beyond that…nichts.

I look forward to seeing more as your work progresses.

Best,

Dave

uuglypher wrote:

Thanks for looking and commenting, Tony.

In the first pair of yours the depths with strongest discernibility are - to my eye- in the middleground (water, curving wall, sw@n) but some, less strongly defined depth intervals extend to the horizon.

In the second pair I see discernible depths as far as the near water’s edge; beyond that…nichts.

Manually revised 2nd image to exaggerate the distant points (Xeyed):-

Yes indeed, the immediate strong perception is of the mass of space between the tree, and shrub at the left and all beyond to the horizon! More subtle yet discernible are depth intervals among the white sailboats to the left of the peninsula projecting from the right.

Whatever you did worked well!  A good start to a productive New Year!

Ciao,

Dave