If you try to pick coordinates in some perceptual space for each of the objects in the experiment then you get tangled up in just the same way that you do with the Penrose staircase: you cannot say whether one object is in front of or behind another one. The solution is to give up trying to assign coordinates to each of the objects.

Interesting interview with Andrew Glennerster about his work on the way we perceive the world as we move about.

Very interesting paper on the efficiency of analyzing hemodynamics (blood flow). 2D schematic layouts and appropriate color palettes improve diagnostic performance over 3D models and rainbow color maps.

We interrupt our scheduled programming to bring you these images from… your brain.

I’m not sure if this has much to do with any of the topics I write about, I could imagine it has some overlap with the perception studies relevant to non-photorealistic computer graphics. Who cares. It’s mind-boggling. And offers shadowish echoes of other research… ie. the brain record of a black women looks caucasian, which would be in keeping with research on empathy and identification – assuming the viewer was caucasian. Fantastic, scary stuff.

How can science become something tangible? By approaching it as a game… watch the above video. Then read the paper and keep an eye on the Lottolab. Below Lotto’s Ted Talk…

I love this kind of exploration. JMS explores reverse perspective. Also check out the silhouette illusion test.

“Painted Russian Orthodox Icons sometimes featured what is called Byzantine-perspective in the buildings and backgrounds – as a way of describing God looking out at the world, through the painting. It’s a beautiful concept; a God’s perspective.”

Thanks Keith!

Steve DiPaola studies how Rembrandt controls the viewers gaze. Fantastic stuff. Thanks to boingboing.

In the discussion round of a recent talk, Ken Perlin discussed the linguistic allure inherent in the following sentence…

Time flies like an arrow,
fruit flies like a banana.

It tingles. And lingers. You have to go back and re-read it. Or you want to hear it again. Why? The words in the second half present themselves in one way due to the repetition, then reveal themselves to be something other than what they were expected to be. A verb becomes a subject, a preposition becomes a verb. They undergo a transformation and there’s a prickly moment in which they exist simultaneously with both meanings.

I’m convinced this is what happens with animation – particularly methods such as hand-drawn and claymation, possibly explaining why animation formats are more likely to be viewed repeatedly. You see something which is obviously a drawn line, or a chunk of clay. But frame for frame it becomes an animal, or a little girl, or a splash. Its simultaneously one thing and another. Its like business built on the rule-of threes; the first iteration establishes, the second confirms and the 3rd rips you out of your expectation, and there’s a key element of timing so that there’s a moment of ‘hang’ where both meanings exist simultaneously. Except with animation, there’s no pay-off moment… just a lingering tingle that accompanies the film.

Its an hypothesis which I’d love to see tested. How might this be done? Well, I’m not sure. I don’t know if anyone’s been brain-scanned while listening to the above sentence, or while watching a comedy routine, but there’s probably a tell-tale flickering of activity. Hook up unwitting college students up and show them a filmed splash and a hand-drawn splash respectively. Then ask them if it tickles.

Images from the incredible Joseph Gilland and here.

Data-mining and non-photorealism are closely related, particularly when it comes to algorithms that abstract data according to variable criteria. They also are disciplines struggling to function under the influence and expectations of the human perception system and its sum-of-all-parts owner, the human. The beautiful mountainous landscape above is visualization of optimism bias as witnessed in earning estimates. The yellow lines represent well-informed daily projections of fairly short-term forecasts. The blue dots represent the arrival of reality.

Head over to the Frontal Cortex by Jonah Lehrer for more insight, including why wall street should be hiring the chronically depressed.

Apparently, soldiers who grew up in either tough urban environments or rural areas with hunting experience both excel at sighting roadside bombs. Sgt. Maj. Burnett explains why these groups have an innate “threat-assessment” ability so much greater than suburban gamers, for example. “Video game enthusiasts are narrower in their focus, as if the windshield of their Humvee is a computer screen.”

In one way, this is non-news; it’s obvious that those who’ve had to rely on subtle visual cues in a real environment will have an advantage spotting such cues. In another way, its a reminder that the graphics we create and the context that we place them in does concretely affect the way we perceive our environment.

MyPenName pointed me to this from the comments section at GurneyJourney.
Related item of interest: Culture and Diverging Views of Social Events by Hannah Faye Chua.

Ed Yong reports on a study by Ibrahim Senay on the effect of grammatic formulation on motivation and performance. Briefly summarized, if you concentrate on the possibility of a task, you are more engaged in the doing of that task than if you concentrate on its certainty. And you’re more likely to repeat that task if its described via the imperfect (“I was doing”) rather than the perfect (“I did”).

Cool. But what’s this got to do with my blog? Well, I can’t help but suspect that this is relevant to the differences in perception between realistic and non-realistic imagery. What better way to envision the difference between the two than as the possibility of something versus its certainty?

.

.

.

.
I’ll leave it to you to decide how relevant this is to computer graphics, but it definitely illustrates the impact that a visual cue can provide – even without consciously perceiving it. Ed Yong writes about an experiment in some detail, in which a political flag was flashed to voters for  such a shortamount of time that they couldn’t consciously perceive it. Do you think it affected their voting choices? You betcha! This is hardly surprising considering earlier experiments about subliminal suggestability, but it is frightening to see how easily manipulated our political and consumption choices are.

Peter Krun Frary encountered this storefront window performer and took some photos. (Encountered here after a tip-off from a friend.) Painted reality often looks convincingly flat because surface texture gives the clues as to contour and depth. Imagine this without the camera’s flash. It makes me wonder – can an NPR solution believably expand into stereoscopic space? We’ve been planning on doing so, at least as an option… but I wonder if the effect might cancel itself out. NPR often strives for a convincing flatness. Thoughts?

I learned that the eye is not like a camera, but more like an extension of the brain itself. I learned that moonlight is not blue, it only appears blue because of a trick that our eyes are playing on us.

The eye isn’t a camera, says Gurney, it’s an extension of the brain. And our brains are deceptive little things. We are quite perceptive to reds and yellows – likely due to the perception of ripeness in fruit. What many don’t know – our traditional color wheel reflects this bias.Wonderfully succinct communicator that he is, James Gurney arranges the hours on a clock with similar bias. As you can see, noon is about 3:30. (Mental not to self: make a biased clock representing the production process with final rendering landing at about 6:45)

James Gurney is not just incredibly talented – the type of talent that reeks of endless hours of having repeatedly and thoughtfully executed every idea – he is very generous with the fruits of his exploration. Click on the clock to go to the fourth post in his overview of the color wheel. (Second mental note to self: BUY HIS BOOK!)

Any professional will – in the course of mastering their trade – learn a language which is more differentiated and precise than that of a layman. This difference can be be painfully experienced when, for example, an artist trained in visual communication deals with a client trained in marketing concerns. Or when a seasoned musician listens to a beginner play bi-tonal midis. Head over to the next-to-last post at cognitive daily. I will sorely miss this blog, as it is one of the few that dealt with perception studies – a key ingredient to understanding the complicated issues surrounding non-photorealistic rendering.

One of the most important aspects of non-photorealistic rendering is also one of the most overlooked. Perception. Over at Cognitive Daily, Dave Munger reports on research by Amy Shirong Lu on race recognition in anime…. result: we assume the character’s race according to our own. Head on over and read the details.

Here you see 3 mammoths… from Prehistoric Park, 2000 BC and Carl Buell. The first two are 3D visualizations and the last is hand-painted. It’s also a differnet species, a Columbian Mammoth… but what’s interesting is the differing intentions of these works. The first intends to be scientifically accurate as well as entertaining. The second just wants to entertain, servicing the image we all already have in our heads as to what these creatures did and dramatically upping the ante.  The third also dramatizes, but it is prime pursuit is accuracy. Of course, the process and budget also play a role: the first two approach the process of generating a visual image as a chain of re-usable assets and are the result of teams of artists. Carl creates this one image by himself, perhaps in dialog with specialists… its volume and form exists as an asset only in his head. A further illustration would be starting all over again, except for this (considerable) knowledge, and an animation would be unfeasible.

Of more direct interest here is the way the first two are presented; as “real” elements within the context of their respective fictions. As such, they represent an actual (albeit fictional) reality – this is what a mammoth looked like, this is what it moved like. You can see people interacting with these creatures, supporting the visual claim.

Carl’s piece is hand-drawn. Despite the high level-of-detail, it displays the artifacts of human creation – zoom in on the larger resolution graciously offered at his site and the hair-covered skin gives way to individual strokes of paint. Carl’s hand is there, and thus we know this is a construction. Even at reduced scale, we feel this fact.

In paleoart, this is an important differentiation. The above Trex portrait by Demetrios Vital inherently communicates that it is an idea… whereas BBC’s Walking with Dinosaurs presents itself as real. I’m always amazed at the amount of information that paleologists gleen off of lumpy, crushed bits mineralized bone and find their sleuthing and debates at least as fascinating as the creatures that they are trying to understand. As an artist, its all too easy to slip into what looks cool, what seems right. I respect the dogma of the scientist who has to restrain these urges in favor of factual support. I also respect those artists and artist-scientists – and there’s an incredible wealth of artistic talent and experience out there!
So, while I love the Walking with Dinosaurs series and what its done for the genre, I eye warily its heritage and hope to present a technical groundwork that will make a constructed ‘hand-drawn’ alternative viable – and bind the many artists into the animated formats. For me, Prehistoric Park has placed itself precariously close to losing its educational ambitions and its necessary to question the goals of the format and the implications of the technical processes used.

It’s great to see this kind of experiential perception work available and hands-on via internet. Follow the image link.

Latest, greatest cg technology catapults cgActors far over the uncanny valley!!!

Keith’sexplorations into toonish 3D has been fantastic, with background design and temporal color flashes. Now he explores boil… okay, he calls it squiggles, but it’s boil. And its in keeping with the whole series… namely, fantastic. Head over and read his own words.

Head over to James Gurney’s Gurney Journey for some great insight into comosition and perception as studied via eye-tracking. It would be great, particularly on an image like the one used, to have tests done in i.e. Japan – as background information receives much more attention there.