I have had a Huawei P20 Pro for a while now, and am amazed by the computational photography that it produces. Newer phones are even better - take a picture and the computer fills in the blanks, whether in the dark, high speed, or just high resolution.

It seems that our reference point, Eyeball MKI may well do the same - our vision isn't as good as we think it is, and the brain takes information from several points, jumps up and down on them and makes an image that we would judge all others by. It fills in the blanks and uses some serious mathematical systems to do so:
https://getpocket.com/explore/item/...-vision?utm_source=pocket-newtab-global-en-GB

 

if you want the details then Google Scholar offers quite some choice of reading. Their 2016 paper is well cited (43 papers) . https://scholar.google.com/scholar?hl=en&as_sdt=0,5&q=Young,+Shapley+and+Chariker&btnG=

 

Thanks for that. It is fascinating that we are designed to make the most of so little info to make an image. Perhaps I will stop slagging off computational photography now

 

Back in the 1960s, the London O and A level biology syllabi had quite a lot to say about eyes. As a budding photographer this appealed to me and was one reason why it was the only science subject I got top marks in (at both levels).

 

Interesting stuff, thanks.

I guess it’s related to the way TV looks pretty good quality while you’re watching it, but if you freeze a frame, you can see that the image quality is very poor. Presumably each frame has gone before our eyes can collect enough information to see the weaknesses.


Chris

 

I'm obviously just guessing but it seems from reading the article that each frame, even with poor image quality would still contain enough information in it for our brain to process and recreate an image. As each frame is rapidly replaced with the next, our brain is recieving tiny cumulative bits of information that allow us to string it all together and see the moving image. Presumably in the same way that we see things moving around us all the time.

 

As, or more, remarkable is when the brain gets it wrong and people hallucinate. It is quite awkward talking with someone who is convinced, because they can see them, some third party in the room.

 

It does make it much easier to believe that someone is actually seeing something that isn't there.
Mind you...It could be that our brain is getting it wrong, and there is someone there

 

My youngest went through a period of hallucinating, but she knew that what she was seeing wasn't really there. We *think* that because she had issues with focusing in one eye, the effort of trying to focus to read meant that the eye essentially 'shut down'. And because the brain wasn't getting signals in when it knew that it should, it made stuff up and put it out there instead. Similar effects have been reported in some people with macular degeneration. We were encouraged to get her playing fast based video games and speedy sports (she chose netball) to get her eye exercising and although she still has some trouble, hasn't had any further hallucinations that I'm aware of.

 

Yeah, well that’s the origin of ghost stories I guess!

 

If you stop to think about the amount of interpretation of our senses that goes on before our conscious thought processes, it's terrifying.

 

It is - but equally, can you imagine how much slower our processing would be if every single thing we looked at was in its entirety sent to the brain for processing? Every movement, every colour, every edge, every gradation of light to dark, every time our eyeball moved? I suppose it's a version of jpeg compression really. The algorithm says 'these greens are close enough, we'll go with this one and you can make up the rest in your head' and our visual cortex goes 'here - ten things you need to know about this right now - you can extrapolate'. It probably goes a long way to explaining how babies develop sight and learn about their environment - and why some thigns are more attractive to babies than other things - start off by determining edges (or whatever it actually is), lock those connections down and then add in the next thing....and on it goes.

 

You’ll have me reading those papers to see how the model works (for Macaques anyway).

 

My original post points to just that - edge detection and then fill in the blanks.

 

The brain will always try to make sense of what the eyes see by progressed learning over years and goes to default. This is an issue in particular with individuals who have become partially sighted who see an object, for example a plant pot sitting on a table, and the brain has to make what use it can of the information degraded vision is providing. The result sometime from a partially sighted person in those circumstances might be, ' Where did that cat come from?' This issue can often be confused with dementia where in fact nothing could be further from the truth. It can be associated with the brain looking for patterns or shapes in a log fire, or alien artifacts on Mars.