results of a new technique for 2D-to-3D conversion

Comments, suggestions?

slings and arrows?

Dave

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

All pairs are affected by a very annoying misalignment.

misalignments? Oh M’gosh!

Howzabout this one?

This is good, but all pictures have vertical misalignment, in multiple areas.

What's the technique?

Karsten Bruun Qvist wrote:

What's the technique?

Hello, Karsten;

two viewes (left eye and rt eye) are made of an en face image. Opposing symmetrical geometric transformations are made of the left and rt-eye views and counterposed appropriately for parallel or crossed viewing. A variety of refinements of the transformations are being studied.

The complaints voiced by Gunnar concerning difficulty foveating vertical (or any non-horizontal) disparities is most commonly expressed by older viewers- mostly over 60-65 years.

Most viewers from teens to 45-50 have no problems with foveation of the range of vertical to horizontal disparities found in 3D conversions of many sorts based on geometric transformations.

The paper by Wormald and Wright on relationship of aging and binocular depth perception is quite revealing in this regard.

Best regards,

Dave

uuglypher wrote:

The complaints voiced by Gunnar concerning difficulty foveating vertical (or any non-horizontal) disparities is most commonly expressed by older viewers- mostly over 60-65 years.

Most viewers from teens to 45-50 have no problems with foveation of the range of vertical to horizontal disparities found in 3D conversions of many sorts based on geometric transformations.

Best regards,

Dave

This assumption is very, very wrong.

The vertical disparities are well known to induce severe discomfort (proportional to the level of disparity), independent of the viewer's age.

Detection of vertical disparity in three-dimensional visualizations

"The desirable value of the above mentioned displacement (vertical disparity) is zero. It means that the image is correct with regard to the disparity issue. If the value does not equals zero(h), the image contains the disparity error. User may properly interpret the three-dimensional visualizations only if the objects from their stereo pair have no vertical displacement. Occurrence of this artifact precludes proper image receiving. The defect has a negative impact on the product quality, but also may results in serious health consequences in the case of longer exposure of the viewer to the discussed anomaly. The only way to avoid such problems is to conduct precise tests of 3D visualizations in order to detect visual artifacts, especially the vertical disparity. Defects connected with the occurrence of the disparity have to be entirely eliminated before releasing the product."

Kirsten, my apologies! Let re-word that first paragraph, Two opposing symmetric  transformations of the original  en face view are made and are the 3D image pair

That’s it! They are masked to comparable rectangles and mounted as the 3D pair.

Davw

3D Gunner wrote:

uuglypher wrote:

The complaints voiced by Gunnar concerning difficulty foveating vertical (or any non-horizontal) disparities is most commonly expressed by older viewers- mostly over 60-65 years.

Most viewers from teens to 45-50 have no problems with foveation of the range of vertical to horizontal disparities found in 3D conversions of many sorts based on geometric transformations.

Best regards,

Dave

This assumption is very, very wrong.

The vertical disparities are well known to induce severe discomfort (proportional to the level of disparity), independent of the viewer's age.

Detection of vertical disparity in three-dimensional visualizations

"The desirable value of the above mentioned displacement (vertical disparity) is zero. It means that the image is correct with regard to the disparity issue. If the value does not equals zero(h), the image contains the disparity error. User may properly interpret the three-dimensional visualizations only if the objects from their stereo pair have no vertical displacement. Occurrence of this artifact precludes proper image receiving. The defect has a negative impact on the product quality, but also may results in serious health consequences in the case of longer exposure of the viewer to the discussed anomaly. The only way to avoid such problems is to conduct precise tests of 3D visualizations in order to detect visual artifacts, especially the vertical disparity. Defects connected with the occurrence of the disparity have to be entirely eliminated before releasing the product."

Hi, Gunner,

Sorry, but you are in serious need of an update in the literature on vertical disparities and binocular depth perception. Here’s a brief selection that will get you started.

Best regards,

Dave

Any misalignment on any axis is immediately difficult to watch.
There is nothing to debate in this regard.
Any professional, or advanced amateur, in the field of stereoscopic imaging can confirm this, which is demonstrated and confirmed by scientific studies.

3D Gunner wrote:

Any misalignment on any axis is immediately difficult to watch.
There is nothing to debate in this regard.
Any professional, or advanced amateur, in the field of stereoscopic imaging can confirm this, which is demonstrated and confirmed by scientific studies.

No debate warranted !  Just get familiar with the research on the topic since the late 1970s.

The  authors of the single article you linked  cited their own publications as 1/3 of their list of references. A red flag.!

The premises of Chas Wheatstone have been accepted folk wisdom for almost two centuries!  Only problem? They were wrong! Time’s up!

you also need to catch up with the literature on the relationshionship of aging and vision - particularly wth depth perception!

Ciao,

Dave

uuglypher wrote:

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Very strangely, I see them better as Xeyed than parallel !

Although elderly and no expert, I make, watch and see many, many 3D images and have no difficulty seeing depth with Anglyphs, Side by Sides and Xeyeds but these leave me unsure and look more 3D to me viewed as Xeyed.

Have you tried viewing them as such? Strange effect, but to me they look flatter Side by Side.

tony brown wrote:

uuglypher wrote:

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Very strangely, I see them better as Xeyed than parallel !

Although elderly and no expert, I make, watch and see many, many 3D images and have no difficulty seeing depth with Anglyphs, Side by Sides and Xeyeds but these leave me unsure and look more 3D to me viewed as Xeyed.

Have you tried viewing them as such? Strange effect, but to me they look flatter Side by Side.

Hi, Tony; yes, I have tried X-view, but the // View works best for me. The images are symmetrically counterposed, and thus intended to work by // view.  Glad to hear that you - as well as I at 81 - are among the 27% of those older than 65  with robust binocular depth perception. I should note that it is not unusual to encounter individuals who paradoxically prefer to X view some image pairs intended for // viewing. No good explanation for it. The ophthalmologists and optometrists I’ve asked about it can only suggest it may have something to do with the poorly understood phenomenon of the “dominant eye”. No big help there!

Here are a few more en face examples to try. Let me know how these work for you!

Best regards,

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

tony brown wrote:

uuglypher wrote:

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Very strangely, I see them better as Xeyed than parallel !

Although elderly and no expert, I make, watch and see many, many 3D images and have no difficulty seeing depth with Anglyphs, Side by Sides and Xeyeds but these leave me unsure and look more 3D to me viewed as Xeyed.

Have you tried viewing them as such? Strange effect, but to me they look flatter Side by Side.

Hi, Tony,

Thanks for your comments. It’s not unusual that some viewers of these 2D-to-3D conversions get better results by crossed gaze technique. I’m glad that you mentioned you are elderly. The study I mentioned on the relationship between aging and binocular depth perception in over 400 people 65 years of age or older did find that 27% had normal (or better) binocular depth perception. Looks like you and I (at 81) have found ourselves in that lucky percentile!

best regards,

Dave

uuglypher wrote:

tony brown wrote:

uuglypher wrote:

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Very strangely, I see them better as Xeyed than parallel !

Although elderly and no expert, I make, watch and see many, many 3D images and have no difficulty seeing depth with Anglyphs, Side by Sides and Xeyeds but these leave me unsure and look more 3D to me viewed as Xeyed.

Have you tried viewing them as such? Strange effect, but to me they look flatter Side by Side.

Hi, Tony,

Thanks for your comments. It’s not unusual that some viewers of these 2D-to-3D conversions get better results by crossed gaze technique. I’m glad that you mentioned you are elderly. The study I mentioned on the relationship between aging and binocular depth perception in over 400 people 65 years of age or older did find that 27% had normal (or better) binocular depth perception. Looks like you and I (at 81) have found ourselves in that lucky percentile!

best regards,

Dave

As a test, I submit 4 (Tri-shot) stereos in both Side x Side and Xeye formats. There should be no ambiguity as to which is the correct pair appropriate to your viewing arrangement. The tri-shot element is mostly very small. As W.S.Gilbert put it: " ... merely to add artistic verisimilitude to an otherwise bald and unconvincing narrative."

tony brown wrote:

uuglypher wrote:

tony brown wrote:

uuglypher wrote:

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Very strangely, I see them better as Xeyed than parallel !

Although elderly and no expert, I make, watch and see many, many 3D images and have no difficulty seeing depth with Anglyphs, Side by Sides and Xeyeds but these leave me unsure and look more 3D to me viewed as Xeyed.

Have you tried viewing them as such? Strange effect, but to me they look flatter Side by Side.

Hi, Tony,

Thanks for your comments. It’s not unusual that some viewers of these 2D-to-3D conversions get better results by crossed gaze technique. I’m glad that you mentioned you are elderly. The study I mentioned on the relationship between aging and binocular depth perception in over 400 people 65 years of age or older did find that 27% had normal (or better) binocular depth perception. Looks like you and I (at 81) have found ourselves in that lucky percentile!

best regards,

Dave

As a test, I submit 4 (Tri-shot) stereos in both Side x Side and Xeye formats. There should be no ambiguity as to which is the correct pair appropriate to your viewing arrangement. The tri-shot element is mostly very small. As W.S.Gilbert put it: " ... merely to add artistic verisimilitude to an otherwise bald and unconvincing narrative."

uuglypher wrote:

tony brown wrote:

uuglypher wrote:

tony brown wrote:

uuglypher wrote:

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Very strangely, I see them better as Xeyed than parallel !

Although elderly and no expert, I make, watch and see many, many 3D images and have no difficulty seeing depth with Anglyphs, Side by Sides and Xeyeds but these leave me unsure and look more 3D to me viewed as Xeyed.

Have you tried viewing them as such? Strange effect, but to me they look flatter Side by Side.

Hi, Tony,

Thanks for your comments. It’s not unusual that some viewers of these 2D-to-3D conversions get better results by crossed gaze technique. I’m glad that you mentioned you are elderly. The study I mentioned on the relationship between aging and binocular depth perception in over 400 people 65 years of age or older did find that 27% had normal (or better) binocular depth perception. Looks like you and I (at 81) have found ourselves in that lucky percentile!

best regards,

Dave

As a test, I submit 4 (Tri-shot) stereos in both Side x Side and Xeye formats. There should be no ambiguity as to which is the correct pair appropriate to your viewing arrangement. The tri-shot element is mostly very small. As W.S.Gilbert put it: " ... merely to add artistic verisimilitude to an otherwise bald and unconvincing narrative."

Tri-shot is just a handy name for what I am experimenting with at the moment and involves taking a conventional stereo pair with lens separation of the order of 65mm to 100mm followed by a single extra shot of about 0,5metres to 3 metres to one side as in hyperstereo Cha Cha shooting. The three images are mounted as layers in Photoshop or similar editing software and the two images from the same side are masked to allow both the normal foreground to be seen, along with the far-spaced background. Thus providing a 'deep stereo' effect.

Alternately, a hyperstereo photo can, by this Tri-shot technique, include a conventional foreground which it otherwise can't, the two foreground images being far too far apart on the resulting picture for the eyes to reconcile together.

In the cemetery scene, I have to say that there IS a 400 metre gap between the rear headstones and the distant tree line. That gap is meant to be seen and there will be no further stereo separation among those distant trees. I don't know about scalene transformation at all so I can't comment?

Why would you want to involve pincushion distortion? I don't see any benefit without a special viewer.

tony brown wrote:

uuglypher wrote:

tony brown wrote:

uuglypher wrote:

tony brown wrote:

uuglypher wrote:

In case y’hadn’t noticed, these are for parallel gaze!--
uuglypher
"100% of the shots you don't take don't go in!"
Wayne Gretzky

Very strangely, I see them better as Xeyed than parallel !

Although elderly and no expert, I make, watch and see many, many 3D images and have no difficulty seeing depth with Anglyphs, Side by Sides and Xeyeds but these leave me unsure and look more 3D to me viewed as Xeyed.

Have you tried viewing them as such? Strange effect, but to me they look flatter Side by Side.

Hi, Tony,

Thanks for your comments. It’s not unusual that some viewers of these 2D-to-3D conversions get better results by crossed gaze technique. I’m glad that you mentioned you are elderly. The study I mentioned on the relationship between aging and binocular depth perception in over 400 people 65 years of age or older did find that 27% had normal (or better) binocular depth perception. Looks like you and I (at 81) have found ourselves in that lucky percentile!

best regards,

Dave

As a test, I submit 4 (Tri-shot) stereos in both Side x Side and Xeye formats. There should be no ambiguity as to which is the correct pair appropriate to your viewing arrangement. The tri-shot element is mostly very small. As W.S.Gilbert put it: " ... merely to add artistic verisimilitude to an otherwise bald and unconvincing narrative."

Tri-shot is just a handy name for what I am experimenting with at the moment and involves taking a conventional stereo pair with lens separation of the order of 65mm to 100mm followed by a single extra shot of about 0,5metres to 3 metres to one side as in hyperstereo Cha Cha shooting. The three images are mounted as layers in Photoshop or similar editing software and the two images from the same side are masked to allow both the normal foreground to be seen, along with the far-spaced background. Thus providing a 'deep stereo' effect.

Alternately, a hyperstereo photo can, by this Tri-shot technique, include a conventional foreground which it otherwise can't, the two foreground images being far too far apart on the resulting picture for the eyes to reconcile together.

In the cemetery scene, I have to say that there IS a 400 metre gap between the rear headstones and the distant tree line. That gap is meant to be seen and there will be no further stereo separation among those distant trees. I don't know about scalene transformation at all so I can't comment?

Why would you want to involve pincushion distortion? I don't see any benefit without a special viewer.

Thanks, Tony, for your explanation of “tri-shot”; it’s a novel and obviously a successful approach!

I used your left eye view as the base image for my 3D conversions. A variety of scalene geometric transformations and paraxial centripetal ( pincushion) distortions can be  applied to copies of the base image or use as the right eye image. The former are used in the case when the perspective vanishing point is beyond the edges of the image; the latter is used to accentuate a vanishing point within the image.

There are, clearly, differences of opinion on the significance of discernible depths in the distant regions of the image! Chacun a son goût!

Best t’ye!

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