What Makes Julian Beever's 3-D Chalk Drawings So Captivating?

[Aether]

Check out these 3-D chalk drawings done by Julian Beever http://users.skynet.be/J.Beever/pave.htm:

 

[zoobyshoe]

Those are stupendous. Particularly that last one.

 

[Mk]

Amazing! I had this guy, I think he is even better! I forget his site and his name! He drew in Italy. Drew for the pope one time I think. Drew things erupting out of the ground and really deep holes. Spectacular.

 

[Mk]

Wow, I found him.

 

[Aether]

Amazing! I had this guy, I think he is even better! I forget his site and his name! He drew in Italy. Drew for the pope one time I think. Drew things erupting out of the ground and really deep holes. Spectacular.

Those really are amazing. What do you mean "I had this guy"? Did you study art under him?

 

[Mk]

Oh, no! I had his site bookmarked and I showed it to my friends. If I was a good artist I would want to have this guy teach me!

 

[Mk]

I can tell you I would not want to find myself walking over one of his drawings. The holes look so real! I wouldn't want to stand on the edge either, on one of those drawn rocks. I would freak out. It looks so real...

 

[Ivan Seeking]

Absolutely incredible!

 

[Physics is Phun]

amazing! It only looks like that from that particular angle right? it's just messed if viewed from the side right?

 

[Gale]

these are really super neat. you can tell they're drawings, but its so hard to believe they're flat. it would definitely be weird to walk over one. its really crazy.

 

[Physics is Phun]

linky?

 

[chroot]

Good ol' trompe l'oil. I assume they'll mess with your head if you view them at a time when sunlight is casting real shadows the other direction.

Related trivia: 90% of normal depth perception is based on information from shadows. Only about 10% is based on real stereoscopic parallax between the two eyes.

- Warren

 

[Aether]

amazing! It only looks like that from that particular angle right? it's just messed if viewed from the side right?

Here's one drawing as seen from the optimal viewing angle:

...and here's the same drawing as seen from a different angle:

These photos are here: http://users.skynet.be/J.Beever/pave.htm
I provided this link in the OP, but it has a spurious ':' at the end of the link, and therefore doesn't work. Would someone with the necessary super-powers please fix that link for me?


My guess is that they are "designed" for optimal viewing from one particular point in space, and at one particular downward angle (negative elevation) in order to allow the viewer's brain to confuse each pixel's longitudinal distance with its vertical height. I suspect that if you were to project an image (e.g., using a slide projector at night for example) from any suitable point in space (preferably at an average person's eye level I would suppose) and at an appropriate downward angle, then you could use a simple "paint by numbers" approach to chalk-down the drawing. Hehe...now, where do I go to get a grant to take a large GPS-guided farm tractor into the desert southwest somewhere and plough some HUGE mega-artwork (maybe an Egyptian pyramid) into the desert so that from an airliner at 40,000ft it looks like there really is a giant pyramid in the desert.

 

[hypnagogue]

Related trivia: 90% of normal depth perception is based on information from shadows. Only about 10% is based on real stereoscopic parallax between the two eyes.

Just curious-- where is this figure coming from? Of course, in addition to shadows and binocular parallax, there are also many other cues we can use to process depth information (e.g. http://webvision.med.utah.edu/KallDepth.html).

 

[chroot]

Unfortunately, I don't have a link. Due to eye-muscle surgery I had as a child, I have only about 10% of normal stereoscopic depth perception. Several optometrists have told me (and my parents) that it doesn't really matter, because 90% of depth perception is based on visual cues like shadows.

It's probably not an "exact" figure anyway, even though it's been repeated to me a couple of times. It's probably more like a rule of thumb.

- Warren

 

[Greg Bernhardt]

yeah chroot is right, I have amblyopia in my left eye, so I see 90% out of my right eye which means technically I should have very little depth perception. However my brain has compensated by using other visual cues and systems. And without much problem since I've played active sports all throughout my life.

 

[chroot]

You're lucky you can play sports, Greg! I'm pretty lousy at anything involving flying balls, so I took up rock climbing and scuba diving and cycling instead.

There's nothing worse than tennis: a small, fluorescent ball, illuminated uniformly by sunlight, high enough in the air that its shadow is impossible to discern. Parallax is the only way to tell where a tennis ball is in mid-air, and I can't do parallax. As a result, I'm probably the world's funniest tennis player.

- Warren

 

[zoobyshoe]

Here's one drawing as seen from the optimal viewing angle:


...and here's the same drawing as seen from a different angle:

That is much more extreme that I imagined. They are pretty much lost from the wrong viewing point. I don't have any idea how they work out what has to be drawn to make it work from the correct angle. There is much more planning to those things than a crop circle.

 

[scott1]

Those are incredible!I wonder if there waterproff.

 

[zoobyshoe]

Just curious-- where is this figure coming from? Of course, in addition to shadows and binocular parallax, there are also many other cues we can use to process depth information (e.g. http://webvision.med.utah.edu/KallDepth.html).

Monocular Cues

Several strong monocular cues allow relative distance and depth to be judged. These monocular cues include:

1. Relative size
2. Interposition
3. Linear perspective
4. Aerial perspective
5. Light and shade
6. Monocular movement parallax

These are very interesting, Hyp, especially the last which I had never noticed, but just checked out and found it to be true: When you move your head things in the distance seem to move slightly in the same direction as head movement. Things close up seem to move opposite to the direction of head movement.

You'd expect the latter, but not the former.

 

[larkspur]

yeah chroot is right, I have amblyopia in my left eye, so I see 90% out of my right eye which means technically I should have very little depth perception. However my brain has compensated by using other visual cues and systems. And without much problem since I've played active sports all throughout my life.

You can still have binocular depth perception with amblyopia providing there is no strabismus associated with it and if the amblyopia is not too dense ( 20/400 or worse.) I have several non-strabismic patients with amblyopia that are able to detect 50 seconds of arc or more.

 

[zoobyshoe]

I have several non-strabismic patients with amblyopia that are able to detect 50 seconds of arc or more.

I'm not sure what this means. Detecting arcs?

 

[larkspur]

Here is an explanation from :

http://www.eyetec.net/group6/M28S2.htm"
"Stereopsis is measured in minutes, or seconds, of arc. This has to do with the angle of separation created by the distance between the eyes and the distance from the eyes to the subject. The smaller the angle is, the more difficult it is to perceive stereopsis, and the better the patient’s ability is. That is, 80 seconds is better than 140 seconds. Results should be recorded in seconds of arc if possible. Alternatively, if one type of test is used exclusively, you can record the results as the number correct, e.g. 6 of 9 circles."

 

[larkspur]

yeah chroot is right, I have amblyopia in my left eye, so I see 90% out of my right eye which means technically I should have very little depth perception. However my brain has compensated by using other visual cues and systems. And without much problem since I've played active sports all throughout my life.

Do a Google search for Babe Ruth and amblyopia...

 

[zoobyshoe]

Here is an explanation from :

http://www.eyetec.net/group6/M28S2.htm"
"Stereopsis is measured in minutes, or seconds, of arc. This has to do with the angle of separation created by the distance between the eyes and the distance from the eyes to the subject. The smaller the angle is, the more difficult it is to perceive stereopsis, and the better the patient’s ability is. That is, 80 seconds is better than 140 seconds. Results should be recorded in seconds of arc if possible. Alternatively, if one type of test is used exclusively, you can record the results as the number correct, e.g. 6 of 9 circles."

I still don't understand. It's not clear to me what the fly and other images are all about. Is it some kind of 3-d illusion? I understand how angles are measured but don't understand what angle is being measured here.

 

[larkspur]

Yes, they are a 3-d illusion created with polarized filter glasses.

 

[Anttech]

yeah chroot is right, I have amblyopia in my left eye, so I see 90% out of my right eye which means technically I should have very little depth perception. However my brain has compensated by using other visual cues and systems. And without much problem since I've played active sports all throughout my life.

I have the same.. And Played sport to quiet a high level actually

 

[Kerrie]

Did a search for amblyopia, and found this thread. Love these chalk drawings! Greg, my youngest also has amblyopia, and I often wonder what her perception is when looking at 3-D images.

 

[Greg Bernhardt]

Did a search for amblyopia, and found this thread. Love these chalk drawings! Greg, my youngest also has amblyopia, and I often wonder what her perception is when looking at 3-D images.

I've been able to see 3-d images just fine and even movies. My mom also has monovision and can also see 3-d movies.

 

[rewebster]

reminds me of Holbein's painting, painted in 1533---there's a hole in the frame that if looked through shows the image (the elongated shape on the floor) in a different perspective:

http://en.wikipedia.org/wiki/The_Ambassadors_(Holbein )

(shows the image about 1/3 down the page on the right from looking through the hole)


you can also 'see' the image if you put your face about an inch or two away from the screen in the upper right corner (about 2 o'clock) and look down at the image--(or up from the lower left corner of the monitor--from about 8 o'clock)