Question for the experts !!

[Gregg MacPherson]

..... there are micro-jitters that an eye makes, .....One article I found calls these "micro-saccadic" motions and "drift". Those called drift have an oscillation not much bigger than the size of a cell. ......

 

 

I think we both enjoy being smart asses sometimes. Don't worry.

 

Tiny angular movements of the eye such that the retina has effectively linear movements a bit bigger than the size of a cell. really interesting. I wonder if the Aaton designers consciously knew of this.

[Carl Looper]

I'd suggest Dennis wasn't necessarily accusing me of making a link between jittering sensors and jittering eyeballs, but was heading off any such assumption, whoever might make such.

 

C

[Carl Looper]

 

I think we both enjoy being smart asses sometimes. Don't worry.

 

Tiny angular movements of the eye such that the retina has effectively linear movements a bit bigger than the size of a cell. really interesting. I wonder if the Aaton designers consciously knew of this.

 

We've yet to establish whether there is anything in such micro-jitters of an eyeball that correspond to the problem being solved by a sensor doing so.

 

The proposed micro-jitter of the sensor is already a solution to a particular problem - that of capturing information otherwise missed by a regular grid sensor. What one might loosely call "in-between" information. Interestingly the jitter doesn't have to be sub-cell in size. Any multiple of such will be fine, so long as it has a sub-cell offset. More technically one wants it to be out of phase by some random offset.

 

C

[Carl Looper]

An auxillary proposition (not proposed by Aaton or Dennis) is to also jitter the architecture of the cell arrangement. And this would be quite definitively bio-mimicy insofar as it is directly inspired by a study of retinal cells. However it's not certain whether the same problem is being solved in both cases - but that doesn't matter because the problem is not how to make a better eye (to play God) but how to make a better movie camera (to play human).

 

To decode the resulting data requires only a one off record of the sensor cell distribution and a dynamic record of the jitter vector (angle, magnitude, or x,y).

 

That decoding can be done for display of the image on a regular grid screen, or in a more physical way: using a custom projector which jitters it's light transducer (in the opposite way) to decode the image.

 

C

[Carl Looper]

Certain physical experiments can be easily done to get a sense of this. Be it as R&D for a better movie camera, or just as an experiment in it's own right. One doesn't have to be pursuing a better movie camera in doing such experiments.

 

A simple one is putting a mirror in front of a camera and jittering it by random amounts, and then projecting the results via a similar mirror that jitters the projection in the opposite way. Some finesse will be obviously required to ensure appropriate syncronisation. One will want to do the jittering in the blanking interval to avoid motion blur.

 

C

[Gregg MacPherson]

There's a nigling doubt..The Aaton vibrating sensor. I don't think it was designed to produce images for a sympathetic vibrating projector. All are clear on that? Or was I wrong.

Edited March 1, 2015 by Gregg MacPherson

[Carl Looper]

Yes, that's right. There's no need to require a reciprocal projector and Aaton would certainly not have been requiring such.

 

C

[John Shell]

On 2/25/2015 at 2:20 AM, Carl Looper said:

 

Yes, I've proposed the same thing in the past. Jitter the sensor. And resample (or jitter the projector - great idea). But as your experiment demonstrates it's still not quite ideal.

 

But it can be improved without ending up back with film.

 

If the sensor cells are randomly arranged in a fashion similar to an organic retina, this should have the effect of neutralising any global correlations that might interfere with image perception.

 

So a sensor that looks like this (a monkey retina), and peizo jittered. And the projector performing the inverse jitter (great idea), or the data can otherwise be re-sampled for a regular grid projector.

 

http://imgc.allpostersimages.com/images/P-473-488-90/72/7273/M3KT100Z/posters/sinclair-stammers-light-micrograph-of-monkey-retina-high-power.jpg

Dunno if it's the same as jittering but the Pentax K-3 dslr had a feature where the sensor gets micro vibrated at a rate of 500hz to produce some blur. This was their method of reducing aliasing without using an anti aliasing filter. Not sure if any other cameras ever had this feature tho

Edited August 27, 2023 by John Shell