Re: New 3D lens announced

Bill G

I am impressed with the determination for making new 3d lenses....
I wish I could jump in, but too slammed with other life endeavors.
I have followed some of the thread...
A few insights that might be helpful.
The goal of using a standard camera and achieving two stereo images is appealing to many of us.  It allows use of current camera technology and solves the lens sync issue better than anything, except maybe for genlock.

The downsides of using a mirror adapter on a single 2d lens.

1)  the captured aspect ratio is not ideal, but with added prisms, this can be helped significantly, as seen in the TriDelta splitter from the 1970's.

2)  The effective f stop is greatly increased (as for aperture diffraction, but not light), as the fl remains the same, but the half aperture creates a higher effective f stop.  A dual lens mirror / prism system solves this issue.

3)  The camera fl must be a bit longer than normal as you can not fold the ray bundle with mirrors without vignetting.

4)  A 2d lens is designed for the incoming ray bundle to enter the spherical front element in a specific spherical format, both horizontal and vertical.  When adding two 45 deg mirrors to a half a lens, the entire incoming ray bundle is now parallel.  A 2d lens was not designed for parallel rays entering the front spherical surface.  Distortion is just one outcome of this mismatch. The amount of distortion depends on the 2d lens design.  An optical ray trace from subject to target (sensor) must be utilized to determine which lenses are less problematic for this use.   The old lens splitters Francois referenced solved this problem by utilizing 2 lenses specially designed to accept the parallel ray bundle.  It's hard to fathom that the recent Canon patent discussed on this forum is the first advancement since those 1940's mirrored approach Francois referenced.  Canons designed solved all these optical issues, however, the aspect ratio issue might still exist, although I suspect they will flip the images with prisms to end up with two horizontal images vs. vertical. 

Outside of the future canon stereo splitter lens, the only other modern approach I have considered to this problem is using dual 2d camera lenses (interchangeable), which project on a fiber optic plate, then onto prisms which brings the dual imagery into the same aspect ratio of the sensor to prevent any pixel waste.   On the 2d camera, you photograph the fiber optic plate with a marco lens at proper magnification to assure the fiber optic plates contents ends up on the full sensor with close to zero waste.  While this is a very feasible but high tech solution, the cost of the fiber optic plates is extremely high...then add in optics, housing, prisms, design cost, etc., the product becomes ultra costly without high volume to amortize the costs.  Since Canon is more experienced with optical elements and prisms, their design is the best chance of a modern version of the 1940's 3d splitters.

BTW, a line diagram is not an optical ray trace.  So it will not determine the distortions which will occur by changing the mirror angles.  The 2d lens design is half the variable which is being left out of the equation. 

We do see how incredibly effective bringing two images from far apart to closely spaced can be, just pick up a decent pair of binoculars for proof.  Most of this technology is within reach of a company that specializes in imaging optics.  This is basically what Canon is intending to build, hopefully.  Just hoping if they do produce such, it does not have a major flaw, such as not flipping the images, and we end up with tall images vs. wide. 

Bill G

On Sat, Apr 10, 2021 at 1:18 PM Antonio F.G. via <> wrote:
On Fri, Apr 9, 2021 at 03:09 PM, Depthcam wrote:
> By the way, such mirror design could produce parallel shots if the angle of the mirrors were adjusted for that.
Actually, no.  This will only work if a  mirror adapter is mounted on a set of two lenses  - as is the case with the Leiz Stemar or the Zeiss Stereotar C.  When you mount a mirror adapter on a single lens and keep the optical axes parallel, each side is viewing one side of the scene - just as if there were no adapter. In order for a left and right image to be recorded of the same scene, the optical axes MUST be converged.
I was not sure about this, but nothing like making a drawing of a "perfectly" parallel mirror adapter, with mirrors exactly at 45°:

The rays from a point at infinity arrive in parallel to the mirrors, and they send them to the lens in parallel as well. Therefore the lens prints a single point in the sensor. This is exactly the same as if the mirrors were not there.

It seems that the way to get two separate L&R images of the point is to deviate the mirrors somewhat away from 45°.
But I am still not sure whether this implies necessarily convergence of the two optical axis (may need another drawing:-)


Join to automatically receive all group messages.