News This Raspberry Pi transparent display is made using a glass dome

[Admin]

VEEB Projects has put together a cool transparent Raspberry Pi display using a glass dome and a program that replicates the Pepper's Ghost effect.

This Raspberry Pi transparent display is made using a glass dome : Read more

 

[bit_user]

If you want a holographic display, these are the real deal:

3D spatial memories, beyond the headset

Combining the magic of 3D holographic displays with the power of generative AI.

lookingglassfactory.com

They use lightfield technology, which involves placing a microlens array atop a conventional LCD panel. The effect is like a real hologram, where the image changes as you move around and the sense of depth can be simultaneously experienced by multiple observers and without any special glasses.

 

[AnimeMania]

Question for someone who understands how this illusion works. If you cut the reflector into the shape of a cone and inserted it into the glass dome, would that create an image that could be viewed, no matter where you are, when looking into the dome? I imagine the image might be curved when viewed, but there is probably a way to compensate for that by manipulating the image.

 

[bit_user]

Question for someone who understands how this illusion works. If you cut the reflector into the shape of a cone and inserted it into the glass dome, would that create an image that could be viewed, no matter where you are, when looking into the dome?

It seems to me like you'd be able to see something from every direction.

I imagine the image might be curved when viewed, but there is probably a way to compensate for that by manipulating the image.

I think the correction would be dependent on what direction the viewer is looking at it from. If you just wanted to show some sort of repeating pattern and didn't mind if it curved off at the sides, then I think that would probably be doable.

 

[edzieba]

If you want a holographic display, these are the real deal:

3D spatial memories, beyond the headset

Combining the magic of 3D holographic displays with the power of generative AI.

lookingglassfactory.com

They use lightfield technology, which involves placing a microlens array atop a conventional LCD panel. The effect is like a real hologram, where the image changes as you move around and the sense of depth can be simultaneously experienced by multiple observers and without any special glasses.

The Looking Glass displays are not lightfield displays, but high density autostereo displays. The image will change with left-right head motion, with enough angles available for the left and right eye to receive the different (and perspective correct) image, but since the panels use cylindrical lenses, moving your head orthogonally to the lenses will have no effect on the displayed image. The trick they use is they slant the cylindrical lenses slightly respective to the panel, so the non-changing axis is just off of vertical so less noticeable and pure vertical headmotions will get a handful of view changes to mask the trick.

On top of that: a lightfield display is not just a 2-axis autostereo display, but a display that reproduces ray angles rather than discrete view segments. Unlike an autostereo display where focus is fixed at time of render, a lightfield display would allow a viewer to refocus on different scene depths and receive the correct image.

 

[edzieba]

Question for someone who understands how this illusion works. If you cut the reflector into the shape of a cone and inserted it into the glass dome, would that create an image that could be viewed, no matter where you are, when looking into the dome? I imagine the image might be curved when viewed, but there is probably a way to compensate for that by manipulating the image.

Pepper's Cone is exactly that.

Like any Pepper's Ghost effect display, the view is orientation dependant, so unless you dynamically update the image with view motion (or display motion) there will only ever be one 'correct' direction where the display shows the desired view.

 

[bit_user]

The trick they use is they slant the cylindrical lenses slightly respective to the panel

Source?

Regardless, it's cool tech and what I said about it is remains true - even if the viewer-dependent 3D is primarily limited to one axis.

a lightfield display is not just a 2-axis autostereo display, but a display that reproduces ray angles rather than discrete view segments. Unlike an autostereo display where focus is fixed at time of render, a lightfield display would allow a viewer to refocus on different scene depths and receive the correct image.

Isn't the difference just in how you render it? If each pixel encodes a matrix of different viewing angles, then it's up to you whether you send it bundles of rays collected by path tracing vs. different views formed by doing fixed or infinite depth-of-field renders.

 

[husker]

Just to be clear: The tech described in this article should in no way be confused with holograms - nor do the authors represent it as such. It's just a display - no more (or less) like a hologram than the monitor you're using to read these comments. The only difference is that they are reflecting the display image off of a transparent surface to make it semi-transparent. Something like this is cheaper and easier, and at least has some very crude hologram effect: Turn Your Smartphone into a 3D Hologram

 

[edzieba]

Source?

Here's a breakdown. They do some funky subpixel rearrangement too, mostly to minimise weird chroma moire issues from having the lens grid and pixel grid misaligned as they would get from a vertical RGB stripe.

Isn't the difference just in how you render it? If each pixel encodes a matrix of different viewing angles, then it's up to you whether you send it bundles of rays collected by path tracing vs. different views formed by doing fixed or infinite depth-of-field renders.

Kiiiiiinda. You could use a very dense autostereo display panel as a lightfield display, but would need to drive it differently. Autostereo displays squidge multiple rendered viewpoints through a fixed plane into a single panel then split those back out again via lens arrays (or aperture masks etc), whereas lightfield displays skip viewpoint intermediaries and figure out what a singular particular virtual ray from the display would be sampling within a scene and what the controbution of that ray to the overall Light Field (hence the name) emitted by the display would be.