Even though there's no shading, a bad resolution and zero context anchoring it in a familiar setting you immediately see it as a 3D object when effectively that's just 3 2D quadrilaterals changing colors and shape. We really need very little to be able to extrapolate depth information from a very limited 2D-only display.
If you did that with 4D cues in 3D space (say in VR or something) I think it wouldn't work. It'd just look like a 3D object changing shape, not a projection of a higher dimension (the same way an hypothetical 2D being would see 3 quadrilaterals changing shape in the above video, not a solid 3D object that would be an abstract concept to them, not a familiar reality).
Could we teach our brain to "think in 4D" and interpret these cues differently? To a certain extent probably, maybe if you spent hours and hours and hours in a VR simulation with 4D objects you'd start getting a feel for it. I doubt you'd ever get as good as with 3D objects since we've been exposed to these literally every waking moment since birth, but maybe I underestimate the plasticity of our brain.
That actually leads me to an other question: is 3D somewhat hardcoded in our brains or is it purely learned? If we were mad scientists and took a baby brain into a 4D "plato's cave" style simulation, could it grow into being able to perceive 4D as intuitively and effectively as we do with 3D space? Also unrelated question: does anybody have a baby I could borrow?
Good lord! Has anyone tried?
This reminds me of the backwards bicycle video on youtube (https://www.youtube.com/watch?v=MFzDaBzBlL0). I... really want to try what you're suggesting.
"So here's why I did. It was a personal challenge. I stayed out here in this driveway and I practiced about 5 minutes every day. ... after 8 month it happened ... in two weeks he (the son) did something that took me 8 months to do".
Well, if you were serious about learning a skill you wouldn't just do one 5 minute session per day for eight months.
A more appropriate training regime would be way more intensive than one 5 minute session per day. Are we to believe he limited his son to one 5 minute session a day?
Do children really learn languages quicker than adults? By age 3, children will probably have words for almost everything. Babies might even say mama and dada by 6 months of age.
Children have several language learning advantages over adults: complete immersion; it is imperative they learn; effectively unlimited time; no responsibilities.
But, it takes them years to learn their native language to basic competency.
Compare: an English speaker can learn Afrikaans, Danish, Dutch, French, Italian, Norwegian, Portuguese, Romanian, Spanish, or Swedish to General Professional Proficiency in Speaking and Reading in 600 hours carried out over 24 weeks (25hrs per week). Cantonese, Mandarin, Japanese, Korean, or Arabic will take 2200 hours over 88 weeks (25hrs per week).[1]
Is it easier for a child? Yeah probably, they don't even have to look after themselves.
If I could live as a child in a foreign language house in a foreign language city, with only two tasks: learn to speak and read the language to basic competency, and learn to ride a backwards bicycle, I'm firmly of the believe I could out pace the 95th percentile of children at both tasks.
As an aside, he says his son is the closest person to him genetically, but aren't his parents both equally as close to him genetically as his son?
1. http://www.effectivelanguagelearning.com/language-guide/lang...
And any siblings.
So it only took him about an hour, over three days, to get it worked out well enough to not immediately fall off, and an hour and a half over four days to get to 50 meters.
There's a comment in on the YT video from Destin saying "the RC helicopter pilot was able to learn it in about an hour but I don't think his brain is normal" - good point, the RC helicopter pilot has more experience understanding reversed input while the helicopter is flying toward him, also he had the opportunity to watch Mike learn first.
I reckon the learning process could be sped up even more by taking the pedals of, turning it in to a balance bike, learning how to ride it down slight inclines, get that sorted then putting the pedals back on.
It’s possible to represent n+1-d scenes into n-d scenes.
This is an old philosophy debate about wheather if I blind person could suddenly see if they’d understand that a sphere is round just by looking at it.
When some people that had had cataracts for 50 years were treated they not only couldn’t tell that a sphere was round, they couldn’t understand shadows or depth of field/distance. They thought shadows on people were black splotches and when something was moving away that it was actually getting smaller.
I suspect visual input gets trained on our neural network like anything else, though we do have some specialized hardware for it.
Actual 3D-sight might include us seeing the inside of objects, what's behind them, etc.
We're aided in converting our 2x 2D sight into 3D mental models through the use of color and parallax. In projections on images, it's easy to just use color and get rid of parallax, and still get sort of close.
Suppose we tried to make a 3D projection of a 4D world, we don't actually get any significant new tools. We're still going to be limited by having to use parallax and color. This doesn't allow us to add enough information to add that 4D projection into 3D as neatly as the 3D into 2D works. At best you could try to distribute the information somehow. For example by giving information about the third dimension through brightness and the fourth dimension through hue. It certainly wouldn't work nearly as well though.
When I was in college I was fairly decent on the ping pong table, call it high amateur. I also have very very bad vision - -5 diopters in one eye and -6 in the other. At the time I wore contacts and I was also really really poor. At some point I lost one of my contacts and for six months I was effectively blind in one eye for the purposes of ping pong. It took me about a month, but fairly quickly I was playing at my old level without anything close to stereo vision.
It turns out that stereovision is the least important of several depth cues humans use, and is only really effective out maybe 10 feet or so--beyond that the parallax is too small to give much useful info, and we rely on apparent size, relative motion, surrounding context, etc. I can catch a ball just fine, as an adult.
It's really only an issue within arm's reach, and then only when I'm distracted. Occasionally I'll reach for something without paying attention and miss by, like, a foot. Also, I can't use 3D glasses. (With the new polarized kind I can at least wear them and see a normal 2D movie. With the old red/blue kind, everything would be either red or blue, depending on which eye I was using.)
You might have an advantage with not being able to watch some 3D movies - you can tell your friends/family that a medical condition prevents you going to them.
I have comparatively uninteresting/normal vision (biggest issue shortsightedness) but I have this too. My left eye sees more blue, my right eye more red. Looks like the cone distribution wasn't perfect.
I'm not sure, but I think the sun's rays don't have much of a blue component: my left eye gets fractionally less sore from bright sunlight, so if it's really bright out I'll generally be closing my right eye.
Once I got the new glasses I was astonished at how flat and "PlayStation-like" everything looked. My depth perception was thrown off and everything looked much closer to me than it actually was.
I think my brain was using blurriness as an indicator of depth, in addition to binocular vision, and with my myopia corrected it lost that information and had to readjust to relying primarily on binocular vision to gauge depth.
Most of us do. But people who only have sight in one eye have no difficulty perceiving three dimensions. And some fairly large percentage of people with binocular vision are stereoblind, typically without even realizing it.
If anything, I'd say it's not that
> We're aided in converting our 2x 2D sight into 3D mental models through the use of color and parallax.
so much as that we're aided in converting our 2D sight into 3D mental models through the use of stuff like color and parallax through the use of stereopsis.
Those without binocular vision can have a fairly large range of non-specific symptoms and perception problems. There is an interesting discussion of the change in understanding in this link. http://www.strabismus.org/all_about_strabismus.html
Heinz von Foerster’s 1970-1971 experiment at the Biological Computing Laboratory for apprehending the fourth dimension is unique ... combining four dimensional geometry, stereoscopic vision, and joystick manipulation of objects on the screen. ... The fourth dimension was chosen as the knowledge to be acquired because there was no chance that any subjects would have attempted such knowledge before the experiment. By allowing the physical “grasping” of the visual object, where one hand coordinated movement on three axes in the 3rd dimension, while the other similarly controlled three axes of movement in the 4th dimension, subjects were able to intuitively figure out that the strange succession of transforming 3D objects they were seeing (with 3D glasses) were cross-sections of a single 4D object.
http://psychedelicsandlanguage.com/dimensionality/
Disclaimer: I studied with von Foerster years ago, and he'd mentioned this. I only today found this online reference. I want those 4D Toys. Steam, here I come.
and even better, this: https://youtu.be/dy_MUfBuq2I
Here is a good intro to 4D Visualization http://eusebeia.dyndns.org/4d/vis/vis
Now I see this on HN :)
So I searched and found this: http://www.urticator.net/maze/
It seems to be exactly what you propose, a 4D world rendered into 3D space. It even has a kind of "stereo" mode, enabling a 3D experience on a 2D moniter.
And to expand on the idea: Once we become familiar with 4D, we can continue and use it to explore 5D, can't we?
It seems pretty clear that looking at 2D images (and in particular, still ones) is a learned task, like reading, rather than an innate one. Both appear to tie into deep structures in the brain, but both are very recent inventions.
From the perspective of developmental stages described by Piaget, children learn to view in three space primarily with objects within reach (parallax pretty much peters out around the ends of your arms. Once the child becomes mobile, she is able to use semantic understanding to estimate the size of distant objects and get a rough idea of distance. The whole human process of seeing is very different from the way, say, a NN is trained on an image: the whole thing isn't gulped in at once, but we foveate on various parts of the image and assemble / confabulate a whole. You can see this in the structure of Chinese classical painting or pre-persepctive European paintings: distant objects aren't sized in any way proportional to their apparent size. This really maps more to how much attention you pay to various objects in the scene.
You then learn to map that into a 3D model which I believe (but am not digging up refs this instance, sorry) has hardware support.
Thus the 2D->3D process exploits a lot of learned and innate knowledge and technique that you have already developed. With one exception you haven't any 4D experience. That one exception is temporal data -- we can easily extrapolate from, say, a sphere shrinking and growing. Apart from that, there isn't much to work with.
So a 4D rendering is just more information in a 2D flat screen. Your brain can't process all of this information. So it "can't see it". But it is there.
+dunno if these coordinates translates to a cube or just some 2-3D object but I hope you get the point.
It would look like VR. 4D toys supports VR already. I tried it but it did not advance my understanding further than the 2D version, although using 3D tracked controllers is definitely a big improvement over a 2D mouse for manipulating 3D objects or 3D projections of 4D objects.
E.g. When you look at a picture with something that looks like a chair, you assume that it’s indeed a chair, and then you can estimate its size/pose/etc. But there are infinitely many non-chair shapes that would produce the exact same projection. It’s just that you won’t encounter them in real life, except maybe in trickshots like this: https://youtu.be/SKpOKXAVjGo
https://en.m.wikipedia.org/wiki/List_of_four-dimensional_gam...
I've tried a maze game and a 4D space shooter before and I could never wrap my head around them. I don't know if it was just poor representations or if it was just because my brain is incapable of understanding.
Actually, the 3D view that 2D man does not understand but which we do understand is... still 2D. My screen is flat.
You can use a 2D viewport to render a 3D scene in a way that is natural and easy to understand for us humans: A human watching the 2D scene can very quickly surmise from a glance at the viewport: Which objects are in the scene, and where are they located, in _ALL_ 3 dimensions?
This raises the question:
Can you render a 4D scene onto a 3D viewport such that us humans are pretty good at understanding where every object in the scene is, in all 4 dimensions?
I assume the answer is 'yeah, you can do that'. I wonder what that would look like.
It's complicated of course; our eyeballs are involved and they kinda work in 2D and not 3D; where we humans can casually glance at a 2D viewport rendering a 3D scene for a second and know what's happening, we'd have to walk around the 3D screen rendering a 4D scene in order to even see everything.