Advances in autostereoscopy (Photonics Spectra)

Light and its behaviour and properties
favalora

Advances in autostereoscopy (Photonics Spectra)

Post by favalora »

Hello -

dannybee requested a discussion area about 3-D display technology, so I offer this link. I was asked to write an article for Photonics Spectra magazine about current topics in autostereo. Those of you who are in similar LinkedIn groups as me or on FB probably saw this already, sorry. Anyhow, the link is here: http://www.photonics.com/Article.aspx?AID=50765. They titled it, "Hitting Every Angle with Autostereoscopic 3-D Displays."

-gregg
dannybee
Posts: 642
Joined: Tue Jan 06, 2015 10:29 pm
Location: visalia
Contact:

Advances in autostereoscopy (Photonics Spectra)

Post by dannybee »

ooo is this based on the pinhole paralax barrier screen .. what driver do you use to run the lcb screen...or is ther one I use the Stereoscopic Player .. sounds very interesting
I think my idea for a dot mask would make the sceen viewing 1/4 the size of the normal screen... I think I can mix views with photoshop
Attachments
dot.jpg
dot.jpg (2.47 KiB) Viewed 8736 times
PinkysBrain

Advances in autostereoscopy (Photonics Spectra)

Post by PinkysBrain »

Hadn't seen the random hole thing, very clever ... unpractical like all barrier/lenticular methods, but clever nonetheless.

A couple of things on 3D displays which caught my eye recently :

C'T reviewed the Toshiba 55L2 ... they were none too impressed with the 3D.

I noticed that Seereal is working on a HMD version of their holographic display (although it's old news) ... I think they have the most practical technologies between that and their next gen autostereoscopic concept (a directed light display, with a lenticular behind the LCD and a high resolution 1D array light source behind it which can direct the image to a single eye ... not 100% original, there is some prior art, but they're the only ones even talking about it). Practical from a user point of view that is, maybe not practical as far as manufacturability goes.

http://www.youtube.com/watch?v=2Hivt5YQMiM
favalora

Advances in autostereoscopy (Photonics Spectra)

Post by favalora »

Thanks for that YouTube link, I hadn't seen it. Nor had I heard of C'T (the German computer technology magazine?)

I have met maybe one person who's seen the SeeReal display. I have seen impressive-looking photos of the imagery, and heard their talk(s) at SD&A, but there's nothing like seeing a 3D display with two eyes at the same time and moving your head a little.

g
dannybee
Posts: 642
Joined: Tue Jan 06, 2015 10:29 pm
Location: visalia
Contact:

Advances in autostereoscopy (Photonics Spectra)

Post by dannybee »

very disapointed that they use cheap fx overlays to demo ther 3d efect.....thats so miss leading...(its like ther selling a concept that they haven't profected yet). :( ..readed several papers lens system for 3d screens(monitors) :D .. I have a iZ3d monitor :D ... it works great....also got a sony 3D HD camcorder for 1/3 the price of retail :D .. works good too....wondering with a lenticuar lens on a lcd screen how many angle of views can you get? :?:
favalora

Advances in autostereoscopy (Photonics Spectra)

Post by favalora »

dannybee wrote:very disapointed that they use cheap fx overlays to demo ther 3d efect.....thats so miss leading...(its like ther selling a concept that they haven't profected yet). :(
dannybee - I'm sorry, I don't understand what you mean. SeeReal uses digital holography in conjunction with other, non-diffractive optics to generate a computationally-economical approach to 3-D display. By "cheap fx overlays" are you referring to a lenticular array?

Gregg
dannybee
Posts: 642
Joined: Tue Jan 06, 2015 10:29 pm
Location: visalia
Contact:

Advances in autostereoscopy (Photonics Spectra)

Post by dannybee »

favalora wrote:
dannybee wrote:very disapointed that they use cheap fx overlays to demo ther 3d efect.....thats so miss leading...(its like ther selling a concept that they haven't profected yet). :(
dannybee - I'm sorry, I don't understand what you mean. SeeReal uses digital holography in conjunction with other, non-diffractive optics to generate a computationally-economical approach to 3-D display. By "cheap fx overlays" are you referring to a lenticular array?

Gregg
this video ad is so unreal http://www.youtube.com/watch?v=2Hivt5YQMiM this is bad advertizing (SeeReal Technologies) 8-) you can tell they use fx to do the floating object :lol: at lest in the ad, why dont they show the real thing why do they have to fake it in the ad???....sorry just my fustrations with no being real.. ther by giving the wrong empressions
Dinesh

Advances in autostereoscopy (Photonics Spectra)

Post by Dinesh »

favalora wrote:dannybee - I'm sorry, I don't understand what you mean. SeeReal uses digital holography in conjunction with other, non-diffractive optics to generate a computationally-economical approach to 3-D display. By "cheap fx overlays" are you referring to a lenticular array?
Gregg, the point is that this kind of thing is nowhere near as good as they portray. The video is faked, or, at best, (let's be kind!) a simulation of what they hope to achieve by FX. I've been asked about the SeeReal technology and have been asked if it's possible to make a hologram that can image plane at 150 feet in full colour! No, it's not possible. The basic concept is in error. Every point of the hologram does not contain every part of the image, unless it's a laser transmission hologram because for this to be true the object's (or scenes) area, A, must subtend a solid angle omega at the eye placed at the hologram's plane, such that it's distance from the hologram is:

R^2 = A/omega

As you can see, there's an inverse square law between the total area of a hologram and it's distance from the hologram. The scene in the video shows a linear law of angle subtended against distance and white light (incoherent) reconstruction.

Even then, It's simply impossible for a white light hologram to show an image at several feet in full colour with no dispersion and no aberrations. The video does not show a reconstruction light, but in order to reconstruct a full-colour white light image holographically at the distances they're implying would involve a fourth power accuracy of the reconstruction source (at least!) in terms of angular spread and direction. If the reconstruction is collimated, then no collimated white light source is that accurate at these distances, simply because the collimating optic would have to be fed from a truly point source, ie a source with zero dimensions, less you have diffractive effects. Also, the collimating optic would have to be hyperbolic or semi-elliptical which would make for a very difficult geometry and a very expensive collimator.

What else? Well, the image shows no coma, no astigmatism, no field curvature etc etc despite it being reconstructed in white light in perfect registration at hundreds of feet. A little difficult to believe. In order to image-plane a full colour hologram at hundreds of feet, you'd require the medium to have a spatial frequency range in several thousand l/mm. What material would that be? Can even computer graphics "draw" a structure to the scale of 10's of nanometers? I wouldn't know, not being CG-knowledgeable, but I'd doubt that CG has gone down to 10's of nanometres.

Now, it's possible that it's clever computer graphics, but I'm pretty sure there ain't no hologram there! They keep showing CG simulations and FX simulations, but as far as I can tell, never the actual device. The proof, as they say, is in the diffractive pudding!
dannybee wrote: why dont they show the real thing why do they have to fake it in the ad???
Because they can't .....
PinkysBrain

Advances in autostereoscopy (Photonics Spectra)

Post by PinkysBrain »

They have pics in their papers and quality seems to have improved over the years. (Some of the pics taking at different focal depths to show it's an actual hologram.) Unless you're assuming they're lying of course.

Generally Lenticulars aren't being used to create a continuum of views anymore BTW. Philips did that for a while, but the viewable resolution just gets too low and you just don't get enough viewing angles to make it look good. For instance the Toshiba display C'T reviewed is a 4K display ... but it requires static viewing, it just uses the lenticular to throw left/right images to the appropriate eye. Even then by the time the lenticular gets done with it they're down to 720p viewable resolution. It works for multiple viewers either through face recognition or manual alignment, which is why they have to give up such a huge amount of pixels (a lot won't be able to be used, because they alias to the wrong eye for one or more of the viewers).

Lenticular works better for monitors, single viewer and relatively known location.
dannybee
Posts: 642
Joined: Tue Jan 06, 2015 10:29 pm
Location: visalia
Contact:

Advances in autostereoscopy (Photonics Spectra)

Post by dannybee »

well I dont know about that Pinky...you will have to ask your friend Brain about if that correct or not.....All I know is my sony with a lenticular screen ..is so way cool...cant waint to try a lenticular screen on a monitor :D
Post Reply