Date   

Re: The big problem with 3D photo closups.

turbguy
 

I thought the whole idea about "close-ups" WAS to "make things appear bigger"...

Wayne


2 x 4K headset.

turbguy
 

Right here.

https://www.pimax.com/pages/pimax-8k-series


Re: The big problem with 3D photo closups.

John Rupkalvis
 

It is true that different people do have a different visual response from others.  This is one of several reasons why no part of a stereoscopic image, including near and far, should ever be out of focus, since you never know what part of the image the person who is seeing it will be looking at during any time.  All parts of the image must be critically sharp.  It is impossible to visually bring any image part to focus if it is not already sharp in the image.  When people try to do so, they often get headaches from the eyestrain.  Many 3D movies currently being exhibited have blurry foregrounds and backgrounds, and this is the main reason why people get headaches trying to view them.  That is why so many people are now turning away from 3D, because they do not realize that their headaches are caused by images that are being shot by people who are not qualified to do so.  They do not understand the basics of photography, let alone stereography.  Good photography, even in 2D, always has all image parts sharp.  So-called "bokeh" has been called aesthetic or artistic.  It is not.  It is just an excuse for poor photography that is an eyestrain to try to look at, and is killing 3D.    

John A. Rupkalvis
stereoscope3d@...

Picture


On Sat, Feb 1, 2020 at 1:14 PM <bglick97@...> wrote:
>> The big problem is the differing displacement between near and far objects.

> That's not a problem.  That's what our three-dimensional world is all about ! ;-)   

         It can be a big problem, details matter... With our unaided vision, when viewing a very close subject, our eye accommodation is on the near, making the far unfocussed.  It's un natural for very near subject to be in focus, as well as a very far subjects.  Also, in the real world, when converged on a near subject, the far subject will appear in double, our brains have been trained since birth for this to occur.  When it does NOT occur in a 3d viewer, it's conflicting to our brains, a form of rivalry, i.e. what our brain expects vs. what is delivered.  Granted, its another one of these conflict issues, some deal with it very well, others will not.  Without details on how close the near is vs. the far, hard to make blanket conclusions.

>  The "problem" if you wish to call it that way is that most regular stereo cameras feature lenses that have a narrow FOV, whereas our eyes have a wide FOV.

       There is no relationship to our eye FOV and taking FOV.   Narrow FOV taking lenses and matching narrow FOV viewing lenses, is still ortho.  Wide FOV of the eye is irrelevant.  The eye is a constant in unaided viewing and 3d viewing.  The only restriction of reduced FOV of viewing lenses, is the window frame effect, i.e. we see the same 3d effect as we would have seen from the camera position in the real world, except through a smaller, restricted window frame.  This is a restriction that will prob. always exist in 3d viewers.  Of course, the less window restriction, the more pleasing.  But its not relevant to the OP position.

On Sat, Feb 1, 2020 at 11:41 AM depthcam via Groups.Io <depthcam=yahoo.ca@groups.io> wrote:
Thanks for essentially outlining what I posted on another thread yesterday.  But let me add a few points and comments:


> The big problem is the differing displacement between near and far objects.


That's not a problem.  That's what our three-dimensional world is all about ! ;-)

The "problem" if you wish to call it that way is that most regular stereo cameras feature lenses that have a narrow FOV, whereas our eyes have a wide FOV.


> One can take a closeup and then realign it for comfortable viewing, but that can move the far objects too far apart.


When shooting close-ups with a wide interaxial, it is generally understood that one must limit the distance of the background.  But even then, "stretch" will occur whenf a normal base is used and the images are then placed behind a virtually distant stereo window.


> You can cross your eyes to see things up close but making them go wall eyed is impossible for most people.


It's all a matter of what's comfortable to most people in real life.  For example, most people find uncomfortable crossing their eyes to look at their own nose but generally find an object a foot away easy to converge upon.  The only obstacle in trying to reproduce this with a "normal" stereo camera (one with lenses 60mm or more apart) is the FOV of its lenses.


> Since the two 3D lenses do not converge for close objects, part of the object is cut off on L & R.


You are being too vague in what you are referring to.  That is only the case with a normal base integrated stereo camera having narrow-field lenses.  There have been custom cameras where the lenses or image plane can be moved laterally and, of course, twin rigs made of independent mono cameras that can be converged.  A good example of this are the recent Sony RX0 twin rigs that have a 60mm interaxial but that can also converge on their subject.

The determining factor here is again whether the stereo image recorded may be viewed at the same virtual distance at which it was shot.


> The alternative is to do closeups with lenses that are close together.


It is very common knowledge that one generally needs to use cameras with lenses closer apart when taking close-up shots.  The point has been brought up mainly because George said he did shoot with his RX0 rig at close range - something that is made easier because of the unique small size of the RX0's and the ability to converge a pair of them at closer distances than would be possible with most cameras - all while maintaining a "normal" interaxial.


> The alternative is to do closeups with lenses that are close together.  Cyclopital had a closeup attachment which used mirrors to achieve this with the W3.  A much cheaper alternative it to get an LG Thrill phone.


There have been many such cameras and lenses over the decades.  Looking back in the fifties, both Leica and Zeiss manufactured optics featuring two lenses in a single barrel (Leica Stemar and Zeiss Stereotar-C) and even Nikon produced a stereo Nikkor.  And old timers will remember the Kin-Dar macro stereo attachment designed by Seton Rochwite that he also later manufactured under his own brand as the "Hyponar".  Even the Italian company ISO produced the  Duplex 120 that had a 30mm interaxial. And who can forget the Macro Realist ?

In the digital era, Panasonic produced the well-known 3D1 and Lumix stereo lens and even Hugo and Jeroen deWijs produced their own macro lenses. And of course, most of the first generation 3D phones had lenses with a smaller interaxial - not only the LG Thrill but the Optimus, the HTC Evo 3D, the Sharp Aquos and even a little known Samsung phone that preceded all of them.  It's also been mentioned a few times here that the very recent RED Hydrogen One features such a stereo lens.


> with the subject appearing to be much larger than real life.


I don't find this effect obvious at all in practice.  People are very much use to looking at 2D close-ups that show small subjects much larger than life-size and adding depth to such images just provides more understanding of small subjects.


> Now if you have a 180 view you can take a closeup...


The objective of a VR180 camera is not to shoot close-ups.  The point I was making is that a close range subject can be recorded with such a camera (even though its lenses are spaced 60mm apart) simply because of the extremely wide FOV of its lenses.  Nonetheless, one still needs to apply common sense in that, if viewing a subject at six inches away is uncomfortable to you in real life, it follows that it will be just as uncomfortable viewing it at the same distance in a VR headset.

Francois


Re: Rokit Phone limitations

John Rupkalvis
 

I use the JS Digitech autostereoscopic 3-D phone.  It displays extremely sharp images with superb depth (dual view or multiview).  If you are in the Los Angeles area, I can show it to you.  

John A. Rupkalvis
stereoscope3d@...

Picture


On Sat, Feb 1, 2020 at 6:42 PM Nima <contactnimaz@...> wrote:

The Rokit IO Pro has a lot of issues.  Their videos from their app are fantastic, but that's the only thing that's good about it.  The screen shimmers when in 2D mode, the 3D viewing angle is tiny, the 3D camera is awful, there's no SDK for developing apps, and their app store has no 3D games or apps as far as I can tell.

It's definitely a budget phone, that's for sure.


Re: The future of 3D: VR, AR, Automultiscopic & Lightfield Displays and Images #futureOf3D #viewing

Nima
 

On Wed, Jan 29, 2020 at 12:10 PM, depthcam wrote:
> I also forgot to mention, I recently got Snapchat's new 3D Spectacles!


Amusing contraption but I can't see any application for these beyond the gadget factor.  Nonetheless, it might be interesting to hear more about the specs and actual picture quality.
It's the best first-person(FPV) camera in the world in my opinion.  The resolution is fantastic(more than 1080p for videos and over 1440p for photos), and they're very convenient.  Downside is they can only shoot video up to 1 minute at a time before needing some seconds to cool down.  The other downside is that they only export to Google's VR180 format, no way to get pure SBS out of them unfortunately.  They also have very cool 3D effects you can add to your photos and videos, and generate in-app wiggle GIF-esque 3D Snaps.


Re: The future of 3D: VR, AR, Automultiscopic & Lightfield Displays and Images #futureOf3D #viewing

Nima
 

On Wed, Jan 29, 2020 at 09:13 AM, <vlarexs@...> wrote:
> Full disclosure, I work for a 3D company in my day job.

What is it? Please tell us more!

I work at Leia as a Product Manager.  We make lightfield displays and 3D software, including software that converts 3D SBS photos to lightfield-compatible multiview images.  I personally work on Holopix, which is a 3D image sharing social network.


Re: Rokit Phone limitations

Nima
 

The Rokit IO Pro has a lot of issues.  Their videos from their app are fantastic, but that's the only thing that's good about it.  The screen shimmers when in 2D mode, the 3D viewing angle is tiny, the 3D camera is awful, there's no SDK for developing apps, and their app store has no 3D games or apps as far as I can tell.

It's definitely a budget phone, that's for sure.


Re: Measure Your Pupillary Distance #theory #viewing #howto

Bill G
 

Thx for added panny info!
Will the display be 4k pixels...or 4k pixels after you turn your head to see all of the imagery? 

On Sat, Feb 1, 2020, 6:16 PM Nima <contactnimaz@...> wrote:

I will second the recommendation for the iPhone IPD measurement application.

I actually went to Panasonic's booth at CES and asked a ton of questions about them!  Actually I even have 3D photos and videos of it from my Snapchat Spectacles.

They claim they'll be ~4K resolution per eye.  The current prototypes only have 3dof tracking, which is fine for 3D and media but not acceptable for games or serious VR applications.  They require a wired connection to either a processing pack, phone, or computer to work.  There's no processing on board, just displays and sensors.  That said, they said that they will try to make it 6dof before launching to consumers.  They said they should be releasing sometime in 2020 if all goes well!


Re: Measure Your Pupillary Distance #theory #viewing #howto

Nima
 

I will second the recommendation for the iPhone IPD measurement application.

I actually went to Panasonic's booth at CES and asked a ton of questions about them!  Actually I even have 3D photos and videos of it from my Snapchat Spectacles.

They claim they'll be ~4K resolution per eye.  The current prototypes only have 3dof tracking, which is fine for 3D and media but not acceptable for games or serious VR applications.  They require a wired connection to either a processing pack, phone, or computer to work.  There's no processing on board, just displays and sensors.  That said, they said that they will try to make it 6dof before launching to consumers.  They said they should be releasing sometime in 2020 if all goes well!


Re: Restoring old Cirvil 3D War Photos

John Rupkalvis
 

For restoring stereo pairs of any kind, including stereo cards, the Stereoid software by Pretend, LLC is really useful.  


Older stereo pairs, like Civil War stereo cards from the 19th century, often have faded differentially, yielding a different apparent exposure and/or contrast (dynamic range) between the left and right eye views.   Stereoid will match them exactly, as well as facilitate many other corrections.   When you are working with color images, or colorizing black & white images, it will match the left and right colors exactly.  

John A. Rupkalvis
stereoscope3d@...

Picture


On Sat, Feb 1, 2020 at 4:23 PM David Starkman via Groups.Io <reel3d=aol.com@groups.io> wrote:
David Richardson did a book on "Restoring and Tinting Vintage Images" and also does incredible restorations of Civil War Stereo views. He also colorizes the, but makes them available restored B&W as well as restored color. See http://historyinfullcolor.com/home/


Restoring old Cirvil 3D War Photos

David Starkman
 

David Richardson did a book on "Restoring and Tinting Vintage Images" and also does incredible restorations of Civil War Stereo views. He also colorizes the, but makes them available restored B&W as well as restored color. See http://historyinfullcolor.com/home/


Re: Restoring old Cirvil 3D War Photos

John Clement
 

No crossview, but if downloaded sView can provide any format you wish.

John

 

From: main@Photo-3d.groups.io <main@Photo-3d.groups.io> On Behalf Of bglick97@...
Sent: Saturday, February 1, 2020 12:04 PM
To: main@photo-3d.groups.io
Subject: Re: [Photo-3d] Restoring old Cirvil 3D War Photos

 

Very impressive John!

do u have these in cross view?

 

On Sat, Feb 1, 2020 at 8:33 AM John Clement <clement@...> wrote:

https://www.youtube.com/watch?v=_-_TY7gLBT4

 

If you click on my name you can get a list of all the other contributions to YouTube.  These have been uploaded as 3D, so YouTube recognizes them and does things to them.  You may have to download them to view them.  Chrome works on a Commander tablet, and I think some 3D TVs can view them properly.

 

John

 

From: main@Photo-3d.groups.io <main@Photo-3d.groups.io> On Behalf Of ratkins5 via Groups.Io
Sent: Saturday, February 1, 2020 7:42 AM
To: main@Photo-3d.groups.io
Subject: Re: [Photo-3d] Restoring old Cirvil 3D War Photos

 

Where can these be seen?

Thanks!

Randy


Re: Restoring old Cirvil 3D War Photos

John Clement
 

I don’t know if I can do that.

John

 

From: main@Photo-3d.groups.io <main@Photo-3d.groups.io> On Behalf Of Jim Johnston
Sent: Saturday, February 1, 2020 11:35 AM
To: main@photo-3d.groups.io
Subject: Re: [Photo-3d] Restoring old Cirvil 3D War Photos

 

Thanks for the link, John.  Your show was so interesting I even put on my anagylph glasses to view it. 

Can you provide a link to Yabazam.com?

 

Jim


Re: May, 1: A new era in 3D photography?: Sony multi-terminal era

timo@guildwood.net
 

I have finally got my Sony a5100 rig put back together with my super short plugs. The minimum lens separation is 72mm. I will have to be happy with it.

Timo

Sent from BlueMail

On Jan 4, 2020, at 11:29 PM, "timo@..." <timo@...> wrote:
What's wrong with the Samsungs? Synchronization is less reliable, recording images to the memory card is less reliable, and the camera staying in the selected mode is less reliable.
I have been shooting with my Samsung rig while the Sony rig is being upgraded, and I still love it. The Sonys are better in low light and are all around a better experience, but not by that much. Still, it can be very frustrating to discover that one of your cameras failed to record a single image for a whole morning.

Timo

Sent from BlueMail
On Jan 3, 2020, at 1:13 PM, turbguy < wkarberg@...> wrote:
...and what's "wrong" with a pair of Samsung NX1000's?  72mm interaxial if unmodified, 68mm if you care to trim off two strap lugs?

Wayne

On Fri, Jan 3, 2020 at 10:40 AM DaveJes1979 via Groups.Io <davejes1979= yahoo.com@groups.io> wrote:
That's good news indeed, Timo.  73mm min. base is great, considering the old Fuji W3 was set at 75mm.

I'm more partial to the a6300, although from my image measurements the min. base would be 83mm.  I'd take that small penalty in exchange for the better image quality, flash shoe, and viewfinder.


Re: The big problem with 3D photo closups.

Bill G
 

>> The big problem is the differing displacement between near and far objects.

> That's not a problem.  That's what our three-dimensional world is all about ! ;-)   

         It can be a big problem, details matter... With our unaided vision, when viewing a very close subject, our eye accommodation is on the near, making the far unfocussed.  It's un natural for very near subject to be in focus, as well as a very far subjects.  Also, in the real world, when converged on a near subject, the far subject will appear in double, our brains have been trained since birth for this to occur.  When it does NOT occur in a 3d viewer, it's conflicting to our brains, a form of rivalry, i.e. what our brain expects vs. what is delivered.  Granted, its another one of these conflict issues, some deal with it very well, others will not.  Without details on how close the near is vs. the far, hard to make blanket conclusions.

>  The "problem" if you wish to call it that way is that most regular stereo cameras feature lenses that have a narrow FOV, whereas our eyes have a wide FOV.

       There is no relationship to our eye FOV and taking FOV.   Narrow FOV taking lenses and matching narrow FOV viewing lenses, is still ortho.  Wide FOV of the eye is irrelevant.  The eye is a constant in unaided viewing and 3d viewing.  The only restriction of reduced FOV of viewing lenses, is the window frame effect, i.e. we see the same 3d effect as we would have seen from the camera position in the real world, except through a smaller, restricted window frame.  This is a restriction that will prob. always exist in 3d viewers.  Of course, the less window restriction, the more pleasing.  But its not relevant to the OP position.


On Sat, Feb 1, 2020 at 11:41 AM depthcam via Groups.Io <depthcam=yahoo.ca@groups.io> wrote:
Thanks for essentially outlining what I posted on another thread yesterday.  But let me add a few points and comments:


> The big problem is the differing displacement between near and far objects.


That's not a problem.  That's what our three-dimensional world is all about ! ;-)

The "problem" if you wish to call it that way is that most regular stereo cameras feature lenses that have a narrow FOV, whereas our eyes have a wide FOV.


> One can take a closeup and then realign it for comfortable viewing, but that can move the far objects too far apart.


When shooting close-ups with a wide interaxial, it is generally understood that one must limit the distance of the background.  But even then, "stretch" will occur whenf a normal base is used and the images are then placed behind a virtually distant stereo window.


> You can cross your eyes to see things up close but making them go wall eyed is impossible for most people.


It's all a matter of what's comfortable to most people in real life.  For example, most people find uncomfortable crossing their eyes to look at their own nose but generally find an object a foot away easy to converge upon.  The only obstacle in trying to reproduce this with a "normal" stereo camera (one with lenses 60mm or more apart) is the FOV of its lenses.


> Since the two 3D lenses do not converge for close objects, part of the object is cut off on L & R.


You are being too vague in what you are referring to.  That is only the case with a normal base integrated stereo camera having narrow-field lenses.  There have been custom cameras where the lenses or image plane can be moved laterally and, of course, twin rigs made of independent mono cameras that can be converged.  A good example of this are the recent Sony RX0 twin rigs that have a 60mm interaxial but that can also converge on their subject.

The determining factor here is again whether the stereo image recorded may be viewed at the same virtual distance at which it was shot.


> The alternative is to do closeups with lenses that are close together.


It is very common knowledge that one generally needs to use cameras with lenses closer apart when taking close-up shots.  The point has been brought up mainly because George said he did shoot with his RX0 rig at close range - something that is made easier because of the unique small size of the RX0's and the ability to converge a pair of them at closer distances than would be possible with most cameras - all while maintaining a "normal" interaxial.


> The alternative is to do closeups with lenses that are close together.  Cyclopital had a closeup attachment which used mirrors to achieve this with the W3.  A much cheaper alternative it to get an LG Thrill phone.


There have been many such cameras and lenses over the decades.  Looking back in the fifties, both Leica and Zeiss manufactured optics featuring two lenses in a single barrel (Leica Stemar and Zeiss Stereotar-C) and even Nikon produced a stereo Nikkor.  And old timers will remember the Kin-Dar macro stereo attachment designed by Seton Rochwite that he also later manufactured under his own brand as the "Hyponar".  Even the Italian company ISO produced the  Duplex 120 that had a 30mm interaxial. And who can forget the Macro Realist ?

In the digital era, Panasonic produced the well-known 3D1 and Lumix stereo lens and even Hugo and Jeroen deWijs produced their own macro lenses. And of course, most of the first generation 3D phones had lenses with a smaller interaxial - not only the LG Thrill but the Optimus, the HTC Evo 3D, the Sharp Aquos and even a little known Samsung phone that preceded all of them.  It's also been mentioned a few times here that the very recent RED Hydrogen One features such a stereo lens.


> with the subject appearing to be much larger than real life.


I don't find this effect obvious at all in practice.  People are very much use to looking at 2D close-ups that show small subjects much larger than life-size and adding depth to such images just provides more understanding of small subjects.


> Now if you have a 180 view you can take a closeup...


The objective of a VR180 camera is not to shoot close-ups.  The point I was making is that a close range subject can be recorded with such a camera (even though its lenses are spaced 60mm apart) simply because of the extremely wide FOV of its lenses.  Nonetheless, one still needs to apply common sense in that, if viewing a subject at six inches away is uncomfortable to you in real life, it follows that it will be just as uncomfortable viewing it at the same distance in a VR headset.

Francois


Re: The big problem with 3D photo closups.

Depthcam
 

Thanks for essentially outlining what I posted on another thread yesterday.  But let me add a few points and comments:


> The big problem is the differing displacement between near and far objects.


That's not a problem.  That's what our three-dimensional world is all about ! ;-)

The "problem" if you wish to call it that way is that most regular stereo cameras feature lenses that have a narrow FOV, whereas our eyes have a wide FOV.


> One can take a closeup and then realign it for comfortable viewing, but that can move the far objects too far apart.


When shooting close-ups with a wide interaxial, it is generally understood that one must limit the distance of the background.  But even then, "stretch" will occur whenf a normal base is used and the images are then placed behind a virtually distant stereo window.


> You can cross your eyes to see things up close but making them go wall eyed is impossible for most people.


It's all a matter of what's comfortable to most people in real life.  For example, most people find uncomfortable crossing their eyes to look at their own nose but generally find an object a foot away easy to converge upon.  The only obstacle in trying to reproduce this with a "normal" stereo camera (one with lenses 60mm or more apart) is the FOV of its lenses.


> Since the two 3D lenses do not converge for close objects, part of the object is cut off on L & R.


You are being too vague in what you are referring to.  That is only the case with a normal base integrated stereo camera having narrow-field lenses.  There have been custom cameras where the lenses or image plane can be moved laterally and, of course, twin rigs made of independent mono cameras that can be converged.  A good example of this are the recent Sony RX0 twin rigs that have a 60mm interaxial but that can also converge on their subject.

The determining factor here is again whether the stereo image recorded may be viewed at the same virtual distance at which it was shot.


> The alternative is to do closeups with lenses that are close together.


It is very common knowledge that one generally needs to use cameras with lenses closer apart when taking close-up shots.  The point has been brought up mainly because George said he did shoot with his RX0 rig at close range - something that is made easier because of the unique small size of the RX0's and the ability to converge a pair of them at closer distances than would be possible with most cameras - all while maintaining a "normal" interaxial.


> The alternative is to do closeups with lenses that are close together.  Cyclopital had a closeup attachment which used mirrors to achieve this with the W3.  A much cheaper alternative it to get an LG Thrill phone.


There have been many such cameras and lenses over the decades.  Looking back in the fifties, both Leica and Zeiss manufactured optics featuring two lenses in a single barrel (Leica Stemar and Zeiss Stereotar-C) and even Nikon produced a stereo Nikkor.  And old timers will remember the Kin-Dar macro stereo attachment designed by Seton Rochwite that he also later manufactured under his own brand as the "Hyponar".  Even the Italian company ISO produced the  Duplex 120 that had a 30mm interaxial. And who can forget the Macro Realist ?

In the digital era, Panasonic produced the well-known 3D1 and Lumix stereo lens and even Hugo and Jeroen deWijs produced their own macro lenses. And of course, most of the first generation 3D phones had lenses with a smaller interaxial - not only the LG Thrill but the Optimus, the HTC Evo 3D, the Sharp Aquos and even a little known Samsung phone that preceded all of them.  It's also been mentioned a few times here that the very recent RED Hydrogen One features such a stereo lens.


> with the subject appearing to be much larger than real life.


I don't find this effect obvious at all in practice.  People are very much use to looking at 2D close-ups that show small subjects much larger than life-size and adding depth to such images just provides more understanding of small subjects.


> Now if you have a 180 view you can take a closeup...


The objective of a VR180 camera is not to shoot close-ups.  The point I was making is that a close range subject can be recorded with such a camera (even though its lenses are spaced 60mm apart) simply because of the extremely wide FOV of its lenses.  Nonetheless, one still needs to apply common sense in that, if viewing a subject at six inches away is uncomfortable to you in real life, it follows that it will be just as uncomfortable viewing it at the same distance in a VR headset.

Francois


Re: New Picture Frames

 

Well, *that* didn't work. The whole purpose for zooming in and cropping the scene in post is to get a decent view of the 3D photos and to be able to actually read the map on the wall.
So, I'll have to figure out another way to reduce the "shakiness" effect without killing the "live" experience.

Thanks for the feedback.
I'll keep experimenting with this...


Re: First true 3D camera announced for 2020

Bill G
 

 >  This is when 3D VR 180 comes in.  With such cameras, there is no need to change the interaxial at close range because the 180 FOV allows one to view the resulting image with the same convergence as we viewed the original.

             While VR viewing was initially for gamers, the VR premise solves one of the biggest problems for capture / viewing 3d.   If you capture a half of a sphere, (180 deg) regardless of the viewing FOV, it will always represent ortho.  This is why I hope the new viewers, such as the Panasonic showed here recently, don't feel compelled to increase the viewing FOV to such extremes as 100deg as it causes so many negative issues, such as size, weight, distortion, cost, etc.  Considering the average person is blown away by a MF viewer at 35 deg FOV, achieving 70 deg diag is still quite the accomplishment.  The ability to turn your head and look around, solves a lot of issues. 3d movies are not shot in 180 deg, nor do they get viewed in IMAX theaters at those extremes... again 70-80 deg is commonplace for IMAX and quite impressive.  It would be great if the Panasonic or other similar new viewers attempt to fill this market niche, vs. ONLY trying to please the gamers. 


On Fri, Jan 31, 2020 at 10:04 PM depthcam via Groups.Io <depthcam=yahoo.ca@groups.io> wrote:
> Hey guys, you have never looked at something 12 inches from your eyes? :)


Looking at something at 12" from one's eyes is not the problem here.  In real life, our eyes have a very wide field of view and are looking at the subject where it is really located.

The problem is that when you take pictures at that distance with a "normal" interaxial, you then change those conditions at the viewing stage.  When one is using a normal or longer lens - your foreground subject may end up very close to the border of each frame.

Ah... but you will argue...  There are two good solutions to that:

A) Back in the days of the Realist, we would shoot at close range with a "normal" base and then mount the images in narrow "close-up" mounts.  This allowed us to pull the film chips out so as to place the close range subject behind the window

B) using separate cameras, we can also just converge on the subject so that it is centered in the image frame.

But in both cases, when we view the picture, it is placed at a virtually greater distance than when we looked at it.  And that results in the effect we call "stretch".  Few people really figured out what causes this effect.  I figured it out by taking a pen and bringing it close to my eyes.  I noticed that when I closed one eye and then the other looking at both sides of the pen, I could see more around the pen as it got closer to my eyes.  That actually was the same as if I viewed a longer pen further away from my eyes.  If I take a picture of the pen at the same distance, but then place it behind a stereo window that is "virtually" further away, my eyes will see the pen as longer than it really is.  Needless to say, that effect will apply to any subject.  And whereas it may go unnoticed with a picture of a pen since we may not have seen the original pen and will just assume we are viewing a picture of a long pen, when subjects are recognizable - such as faces - the effect of "stretch" will be disturbing.

That's why it is not advisable to use a normal base at close range unless you are using a camera system that allows you to view the subject with the same convergence and field of view as in real life.  This is when 3D VR 180 comes in.  With such cameras, there is no need to change the interaxial at close range because the 180 FOV allows one to view the resulting image with the same convergence as we viewed the original.

Of course, nothing stops you from breaking those rules for special effects !


> I have used the Sony RX0 rig (60mm interocular) for close-ups in the range of 12-24 inches. If the background is blocked, you can get interesting pictures.


It's true that with very wide lenses, OSD (on-sensor deviation) is not as great.  But you are still likely to get perspective deformations that may not be immediately obvious until you try to shoot people at that distance.  The way around this is to use tele lenses.  But then you will no longer be at 12 inches.


> I personally think that the 90cm close focus distance is a limitation.


It's not ideal, but that's how this lens was configured, unfortunately.  The only way around this would be to use negative diopter lenses.


> Most compact cameras can focus a lot closer than that


But that's because compact cameras generally use longer lenses therefore require a focusing mechanism. With the advent of very wide angle lenses on action cams, many manufacturers simply do away with a focusing mechanism and set the lenses permanently on the hyperfocal distance for the widest aperture.

From what I understand the lens on the Sony RX0 does have a focusing mechanism.  That's what allows it to focus closer whereas with the GoPro, it is the smaller sensor that provides greater DOF.

As I wrote earlier, if one does not plan to shoot 3D videos or 3D 180, an RX0 rig may be a better choice in that price range.

Francois


Re: Restoring old Cirvil 3D War Photos

Depthcam
 

> Can you provide a link to Yabazam.com?


Yabazam doesn't exist anymore as it was part of TriDef, which went bankrupt some time ago.

Francois


Re: Restoring old Cirvil 3D War Photos

Bill G
 

Very impressive John!
do u have these in cross view?

On Sat, Feb 1, 2020 at 8:33 AM John Clement <clement@...> wrote:

https://www.youtube.com/watch?v=_-_TY7gLBT4

 

If you click on my name you can get a list of all the other contributions to YouTube.  These have been uploaded as 3D, so YouTube recognizes them and does things to them.  You may have to download them to view them.  Chrome works on a Commander tablet, and I think some 3D TVs can view them properly.

 

John

 

From: main@Photo-3d.groups.io <main@Photo-3d.groups.io> On Behalf Of ratkins5 via Groups.Io
Sent: Saturday, February 1, 2020 7:42 AM
To: main@Photo-3d.groups.io
Subject: Re: [Photo-3d] Restoring old Cirvil 3D War Photos

 

Where can these be seen?

Thanks!

Randy