Mars Wiggles

August 21, 2012 12:04 AM

Mars Wiggles Enhancing Curiosity's imagery from Mars using the wiggle stereoscopy effect. [via mefi projects]
posted by circular (28 comments total)

Oh, excellent.
posted by BungaDunga at 12:09 AM on August 21, 2012

*turns on some hard techno*
*stares*
posted by cthuljew at 1:28 AM on August 21, 2012

Thanks. Wiggle stereoscopy is a wonderful thing!
posted by Anything at 3:19 AM on August 21, 2012

I wonder if Curiosity might have the means to implement my idea back from the Phoenix mission days.
posted by Anything at 3:24 AM on August 21, 2012

As someone who only sees out of one eye*, and thus has little idea of stereoscopy or 3D what-have-you, is this wiggle stuff actually "neat" for people?

Honestly all I see is a pulsating image of two slightly different viewpoints which just kinda makes everything look squishy or, I suppose "wiggly."

Is there something I'm missing, where this is actually like an innovative way to interpret spacial differences or something? (honest question) Or is it just "haha! Wiggly pictures!"

* my left eye works at about 10% capacity, but my brain more or less totally ignores it unless I close my right eye
posted by ShutterBun at 3:35 AM on August 21, 2012

(hmm...from the Wikipedia link: "It is the only method of stereoscopic visualization for people with limited or no vision in one eye.") I guess I should be grateful. I mean, I get what they're doing (showing how "far away stuff" moves less than "close stuff,") but I wondered whether there was some "magic eye" effect that I'm not privy to.

Working as intended, in that case.
posted by ShutterBun at 3:38 AM on August 21, 2012

ShutterBun, it depends on the image but in the best cases I at least get a surprisingly strong illusion of depth, vaguely similar to what I get with 3d glasses.

It's unfortunate if it doesn't work for you, but presumably those neurons are put to good use for something else :)
posted by Anything at 4:01 AM on August 21, 2012

Out of curiosity, does it work the same with one eye closed?
posted by ShutterBun at 4:35 AM on August 21, 2012

I wonder if Curiosity might have the means to implement my idea back from the Phoenix mission days.

(the idea):

I wonder if they also record the vibrations in the arm that does the scooping. If they do, that would mean that the vibrations could also be 'replayed' on a metal bar, maybe even one that follows the arm's exact movements.

You could place your hand on it and feel what it's like to scoop Martian soil.

That'd make a hell of a hands-on science exhibit. Might one able to construct this from accelerometer data or a microphone attached to the rover's body as it scoops? According to a JPL guy in that Reddit AMA thread, Phoenix did have a microphone but recorded only static and Curiosity has no microphones. Pity.Maye there's not much atmosphere on Mars to carry the sound of interesting things like dust storms, but some kind of acoustic feedback transmitted through the chassis seems like it'd be useful. At least handy for engineers to be able to hear motors whirring to assess their condition and traction perhaps? Heck even some authentic white noise to put behind the HD surface video the MastCam is going to be providing would be fun.

Speaking of Phoenix, I'd love to learn what the engineers and designers really truly thought about the chance of the spacecraft coming back online on after spending the winter encased in ice. NASA kept saying is was entirely not to be expected, a lot of communication attempts sure were made just to humor the silly public. But gosh darn it, they named the thing Phoenix. Someone had dreams of getting to use words like rebirth and resurrection in press releases.
posted by floam at 4:45 AM on August 21, 2012

It gives me a headache
posted by Flood at 4:50 AM on August 21, 2012

Out of curiosity, does it work the same with one eye closed?

I wondered about that too -- for me it does work.
posted by Anything at 5:17 AM on August 21, 2012

Good wiggle stereoscopy gives me the illusion of depth, but most of these aren't very well rendered, unfortunately. Probably because the images from which they were constructed weren't really meant for this. The two points of view are too far displaced from one another in the black and white close ups, don't appear to be displaced at all on the one with the shadow, generate too little parallax in the distance shots of the mountain... And the descendent imager doesn't seem like it's even an attempt at stereoscopy, just a short movie clip.

Only the two color close ups really give any depth illusion to me. And I wish I could also see the two images from which each is constructed side-by-side. I'm pretty good at crossing/parallelizing my eyes to do Magic Eye effects, and would probably get a stronger sense of 3D from that.
posted by OnceUponATime at 5:45 AM on August 21, 2012 [2 favorites]

ShutterBun -- for what it's worth, these particular images mostly don't create a 3D effect for me either, even with two eyes. The wiggle effect seems to work best with two images that are very close together and with wide separation between the foreground and background, like this or this.

If those aren't doing anything for you either, the best analogy I can offer is this video of head tracking with a WiiRemote. The 3D effect once he turns head tracking on is basically the same effect that these wiggle images create.
posted by jhc at 5:47 AM on August 21, 2012

Is this going to become the new tilt/shift and bore the arse off me when every film and TV programme starts using it?
posted by GallonOfAlan at 6:09 AM on August 21, 2012

Enhancing?
I've seen "wiggle" pics that actually do achieve the intended simulation, but these are total crap. The difference between the images used are too great.
posted by Thorzdad at 6:28 AM on August 21, 2012 [1 favorite]

ShutterBun - I also have amblyopia, I imagine most of us lack the ability to sense depth or 3 dimensions. as our brains developed and only processed the image of one eye, something got lost in the wiring. I often get asked what it's like to see in only 2 dimensions, of course i have no idea what 3 dimensions look like, it's like asking a colorblind person what it's like to not be able to tell the difference between red and green. you can't explain a phenomenon you don't experience. so yeah, for the others there is an illusion of depth, for us it's just 2 flickering images.
posted by camdan at 6:32 AM on August 21, 2012

Is there something I'm missing, where this is actually like an innovative way to interpret spacial differences or something? (honest question) Or is it just "haha! Wiggly pictures!"

A friend of mine, as he describes it, has a "lazy eye". He essentially experiences images from only one eye at a time despite them both more or less working, and claims not to experience typical depth perception resulting from binocular vision. He's remarked that the wiggle stuff is pretty close to as good as 3D gets for him. He relies mostly on motion parallax and shading to determine depth, and I think these animations get that relevant data to you. This is just from prior discussions we've had about wiggle-effect animations that are out there, not the ones linked to on this FPP in particular. Also, I'll note many of these have angles too far apart to work, or something else wrong and aren't very effective, probably due to a lack of sequential images available taken at whatever the sweet spot angle is from Curiosity so far. The subtle ones with close angles work best for me. I'd encourage you to audit other wiggle animations unrelated to the Mars stuff that have been super carefully put together under optimal conditions.

With my unremarkable eyesight (however a knack for taking a few minutes to get those "magic-eye" images to converge properly and kick in), while wiggle animations will produce a very strong appreciation of depth and sorta "look 3D", it is not exactly the same experience as the indescribable, very subtle field of instant whatever-ness that I normally experience or 3D glasses would provide that I call depth perception. For me no amount of staring at these will really get a rock to truly "go" to that position "further away" that I can recognize and feel from the wiggling it belongs at, relative to other objects and the dirt. Things don't pop out or pop back. For me these are just better than closing one eye and bobbing my head around but not fundamentally different. Like I said though, I'm slow to get those "magic eye" autostereograms to click in, maybe others experience these more effectively or I'm flat out doing these wrong.

Out of curiosity, does it work the same with one eye closed?

One eye or two eyes doesn't make a difference for me.

Just now, closing one eye to test that, I should remark that after trying to navigate my dark room and looking at stuff strewn across my cluttered desk with my hand still over that eye, I am very unpracticed at getting on without stereo vision. Considering how I seemingly am not doing much with these depth cues, I can't help but wonder if the wiggle animations should tend to actually have a stronger effect on individuals without binocular vision. With more practice using one eye, conceivably their visual cortex might learn a trick or two it wouldn't otherwise to suss out positions and maybe it constructs a similar whatever perception from different data.
posted by floam at 6:49 AM on August 21, 2012

This one works OK. A lot of the other ones are just not a good use of this technique. Or aren't even this technique at all and are just animated GIFs. Feels like a case of "gotta fill the page with something."
posted by smackfu at 7:26 AM on August 21, 2012

girllookaddatmarsdust
posted by cmoj at 8:25 AM on August 21, 2012

This Mars, it vibrates?
posted by Kabanos at 9:01 AM on August 21, 2012 [1 favorite]

I'm not getting any 3-D off of these guys, but the back and forth replay of the heat shield drop is timed perfectly to "Pump Up The Volume", and even fits in with the imagery from its video.
posted by benito.strauss at 9:59 AM on August 21, 2012

Wiggler here. They're not all great wiggles -- especially where the rover's Navcam is involved, due to the unnatural distance between the lenses -- but I'll glean what clean parallax I can get from any paired image on the MSL raw image feed.

And yes, there are cases where I'm GIFing for the sake of the GIF, not actually wiggling, but going for a sense of movement or dimensionality. I try to mark those as "not wiggles" in the captions.
posted by brownpau at 12:04 PM on August 21, 2012 [4 favorites]

Another amblyopian checking in late. Works ok for me.

BTW, I've always been able to draw "in 3d" better than most people, and I have long held a pet theory that this is due in part to my amblyopia, the idea being that I don't get distracted or thrown off by the difference in stereoscopic cues between the real world and a flat piece of paper, because I'm not getting any of those cues. Wondering if other amblyopians also find they are able to draw better than most people.
posted by smcameron at 5:31 PM on August 21, 2012

I know everybody's (justifiably) moved on to the new hotness that is Curiosity, but every time I'm reminded of it it blows my mind that 8 and a half years on, Opportunity is STILL GOING. Wow.
posted by the painkiller at 7:15 PM on August 21, 2012

smcameron - i have read somewhere that amblyopes are great artists for just the reason you describe. no idea where, though.
posted by camdan at 9:53 PM on August 21, 2012

OK, I think I'm getting what I more or less suspected, but didn't really want to say in so many words (since this was a MeFi project and all) but it seems that "wiggle pictures" can kinda give the 3D effect, but that these (for the most part) aren't particularly strong examples of it. Makes sense.

I also wonder about "drawing in 3D" as you describe, painkiller, since essentially everything I see is in 2 dimensions, so it's not so much of a trick to "convert" it to a drawing. Can't really confirm anything other than that I've never really "had a problem with it."

Another uhhh, benefit, would be "not having to close one eye when looking through a telescope" or microscope, although from what I understand, the correct procedure even for stereo-sighted people is to keep both eyes open, and train your brain to ignore the other eye.

But as long as we're on the subject, is it OK if I ask about binoculars in movies? In movies, it is always presented as sort of a "figure 8" type of shape, with two circles which partially overlap, forming a more or less coherent view. People have told me that the "actual" view through binoculars is just a circle, the same as a telescope would be. When I look through binoculars, I get two circles, partially overlapping but not "connected" to each other.

In other words: this.

So tell me, do binoculars really provide simply a circular view?
posted by ShutterBun at 5:05 AM on August 22, 2012

Brief update and thanks.
posted by brownpau at 4:11 PM on August 24, 2012

I think what I see through binoculars depends on how far away they're focussed. If they are weak enough that I can still see things relatively close up, where my eyes still have different points of view, then my over all field of view will still be wider than it is tall, a sort of oblong shape (I don't see any closing in of the circles toward the middle -- my brain happily ignores any black areas in the field of view of one eye that happen to be covered by the field of view of the other eye the same way it normally ignores my nose -- unless I'm looking for it.) If they are focussed so far away that both eyes see essentially the same thing, my brain fuses those two nearly identical images into a single circle --but I'm pretty sure I don't have any depth perception in that case.

With two toilet paper tubes, I'm pretty sure I can vary the angle between them to see single circle, an oblong, a "movie binoculars" view, or two distinct circles, by pointing them further and further away from one another.
posted by OnceUponATime at 5:33 AM on August 25, 2012