Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | General Gear-talk | Join Upload & Sell

  

Archive 2012 · what would be some of the theoretical ways to increase digital dynamic r...

  
 
jrs5fg
Offline
• •
Upload & Sell: Off
p.1 #1 · p.1 #1 · what would be some of the theoretical ways to increase digital dynamic range?


In a 100 years, should I expect digital image representation anywhere near that of the human eye? What are some of the theoretical pathways to increased dynamic range?


Dec 26, 2012 at 11:26 PM
justruss
Offline
• • • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · what would be some of the theoretical ways to increase digital dynamic range?


Not sure about questions 1-- the human eye (and the brain-based, near-live processing of the multidimensional image we get) is totally different from any capture technique and display medium we currently have. Our capture device is 3D itself, and our display device (brain) is abstract, internal, and massively processed.

True representation would probably require using our eyes to capture and our brains to display. Get between these systems so as to store externally, do pre-processing, etc and then pipe back into the brain and you might have that. Imagine a wireless chip installed somewhere in the circuitry between eye and brain that both would allow external capture/storage of images read by the eye as well as transfer for viewing directly to the brain (taking over temporarily the normal vision). Or direct inputs in the brain...

As for increased DR-- which does not equal natural representation (see HDR, though clunky in implementation) shouldn't be very difficult actually. Expensive, sure, but totally doable. I believe crude versions of what I'm going to propose is already used in certain sony sensors. The extreme is being able to control every pixel's sensitivity individually using some kind of algorithm. If these can be adjusted on the fly during exposure, with some input by the user, done. Just tune the "Base" ISO to be somewhere in the middle of the range so that you can get a few more stops in each direction. Or use multiple full-sized sensors each effectively capturing 1 to 2 EV apart the same image. Extend these ideas in both directions, and you get pretty damn wide DR possibilities. Functional and $$$ limits surely get in the way. But those are totally possible to overcome. And it will happen.

It's a bit like the lytro concept for DOF control. We can do it, but it's pricey and clunky. And current imagery is good enough and established enough that there's a serious barrier in front of the funding and timeline to allow these technologies to advance and rule the market in the short term. In the long term... it's all possible.





Dec 27, 2012 at 04:54 AM
BenV
Offline
• • • • •
Upload & Sell: Off
p.1 #3 · p.1 #3 · what would be some of the theoretical ways to increase digital dynamic range?


Easiest way to increase DR is do HDR images.


Dec 27, 2012 at 09:43 AM
AmbientMike
Offline
• • • • • •
Upload & Sell: Off
p.1 #4 · p.1 #4 · what would be some of the theoretical ways to increase digital dynamic range?


DR has been a limiting factor since Lascaux. But yeah shoot hdr.


Dec 27, 2012 at 12:21 PM
jrs5fg
Offline
• •
Upload & Sell: Off
p.1 #5 · p.1 #5 · what would be some of the theoretical ways to increase digital dynamic range?


To me, HDR doesn't really increase dynamic range information, it just /maps/ information for a high dynamic range into an intensity space with less dynamic range. I'm talking about where the dynamic range actually *increases*, not getting around it by tone mapping.


Dec 27, 2012 at 03:51 PM
BenV
Offline
• • • • •
Upload & Sell: Off
p.1 #6 · p.1 #6 · what would be some of the theoretical ways to increase digital dynamic range?


What exactly are you trying to do? Or is it a thread strictly just to converse?


Dec 27, 2012 at 05:02 PM
jrs5fg
Offline
• •
Upload & Sell: Off
p.1 #7 · p.1 #7 · what would be some of the theoretical ways to increase digital dynamic range?


It's a thread mainly to converse, also because I'm a physics and biochemistry student about to graduate thinking about possible future research directions and steps that could be taken in such directions.

(I could continue my current line of work working on fruit flies, dopamine pathways and cocaine, but I'd prefer to find a research position involving science, optics and photography...)



Dec 28, 2012 at 12:20 AM
justruss
Offline
• • • • •
Upload & Sell: Off
p.1 #8 · p.1 #8 · what would be some of the theoretical ways to increase digital dynamic range?


jrs5fg wrote:
It's a thread mainly to converse, also because I'm a physics and biochemistry student about to graduate thinking about possible future research directions and steps that could be taken in such directions.

(I could continue my current line of work working on fruit flies, dopamine pathways and cocaine, but I'd prefer to find a research position involving science, optics and photography...)


Well...

In the short term-- advances that hit consumer photography will mainly come from engineers.

In the medium term-- they'll come from physicists.

In the longer term-- they'll come from bio/neuro chemists.



Dec 28, 2012 at 06:38 AM
gschlact
Offline
• • • •
Upload & Sell: Off
p.1 #9 · p.1 #9 · what would be some of the theoretical ways to increase digital dynamic range?


jrs5fg wrote:
To me, HDR doesn't really increase dynamic range information, it just /maps/ information for a high dynamic range into an intensity space with less dynamic range. I'm talking about where the dynamic range actually *increases*, not getting around it by tone mapping.


Interesting interpretation you post here about HDR. On the surface, you are correct, but ultimately, it is the representation that we perceive isn't it? Arbus there needs to be a means to portray that representation to our eyes/brains. So th remapping is necessary given current limitations of print. Think about it, if your eye had infinite dynamic range with a brain infinite perception ability, in order to translate a high dynamic range real-world image, you would then need a media that would be able to produce the highest and lowest light level. Considering the brightest levels would include something like the sun, you know it will be impossible to reproduce. So ultimately to Actaully Increase dynamic range as you say, it is more about the media to present it to our eyes since there are many current and future HDR technical techniques that will capture the range of the image that can be combined into huge image dynamic range. The trick and limitation is to then present it as such. Print won't cut it as the brightest whites are limited by the intensity of the ambient light reflecting off the media. Thus to increase dynamic range presentation of an image over print, you would need and Active Display with plenty of resolution, fine pitch pixels, and fine high bit control of luminosity.

Ultimately I believe you would rapidly hit the dynamic range limitations of your eye which I don't believe is too much better than current Monitor technology and HDR image creation techniques. (Think about a monitor that is super bright that you would hardly be able to look at.) Think about being on a beach, laying back on a blanket and someone steps almost between you and the sun/sunny sky; you can't see their face of you look at the sky, and viss versa if you block some of the light so you can see their face- essentially you are Remapping tonality so that you can adjust your perception into a range for you to view the detail of interest.

So, now you are left with ways of expanding the humany eye/brains dynamic range.... Maybe it has to do with Stereo Active displays where essentially you could split the high and low, one for each eye. Each display would ideally be able to cover half of the eyes true dynamic range. (This assumes the pupils and optic nerves act independently from each other which I believe is the case You can do this now using two versions of the image that barely overlap in DR. The neuroscience part then would be how to untrained the the brain for normal Single-eye Dominance perception issues, but I do believe this is possible much like military pilots did /do with their HUD displays that blocked part of their vision but needs incorporation as one image to the brain between both eyes. An alternate to the stereo displays should you have one monitor with such dynamic range would be simply to creat the stereo visual split for each eye with passive glasses with clear on one side (pupil adjusts and can properly perceive the brights), and low pass filter (ie sunglasses) for the other pupil so that it opens to see the dark detail. Likely not as good as the dual stereo displays but cheaper :-).

For the science part, I think the research would be interesting to learn truly how much independent eye imagery could be integrated by the brain to expand (double) normal dynamic range possible at any different moment in normal natural scenery. Even if not double, the technology is already there using current display technology to create the stereo two images expanding the Instantaneous DR of a single pupil, and see how much detail is recognized simultaneously by the brain by proving the two (high and low luminosity) images respectively to each eye.

Thoughts?


Guy



Dec 30, 2012 at 12:11 PM





FM Forums | General Gear-talk | Join Upload & Sell

    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.