Talk:Depth perception

From Wikipedia, the free encyclopedia

Contents 1 Discussion1 2 Discusion2 3 Discussion3 4 Monocular cues 5 Philosophical implications 6 Squirrels 7 binocular vision not necessary 8 Proprioception plus Depth Perception? 9 Brain vs. eyes 10 Untitled discussion 11 The name of this article

[edit] Discussion1

1. Oculomotor – sense the position of our eyes and the tension of the eye muscles (only close range) - Convergence Inward movement by near object - Accommodation Changing of the shape of the lense 2. Monocular – cues that work with one eye

     - Prictorial Cues
        Source of deep information...?

- occlusion (one object hides a part of the other) [whole range] - relative hight (higher is futher away -> horizont) [works by close & medium range] - cast shadows (shadow to help to locate the depth) - relative size (by known same size objects -> smaller one is futher away) [whole range] - familiar size (by know differen size objects, same size -> the known smaller one is closer by) - atmospheric perspective (distance objects are less sharp (on the earth -> air)) [long range] - linear perspective - drawing system of linear perspective (duplicates the pricorial depth cues) ??? - depth cue of linear perspetive (due converge of very far lines (imagine: railway tack)) - texture gradients (depth information from the ground)

      - Movement-Produced Cues

- Motion parallax [works by close & medium range] - > far objects move slowly; near objects move rapidly -> same when you look on the Retina (far objects move less) - Deletion & Accretion - > deletion (covered up) & accreted (uncovered) when moving more closely -> related to motion parallax & overlapping 3. Binocular – cues that depend on two eyes [works by close & medium range] Binoocular Disparity and Steropsis!!!

|-> different image of the two eyes     |-> Impression of depth created by disparity

[edit] Discusion2

The page doesn't provide any information about why depth perception fails, or why people with failed depth perception experience doubled vision when viewing 3d animated images designed to be viewed with shaded glasses with such glasses Hackwrench 04:56, 27 September 2005 (UTC)

As you wish. I've added a blurb on amblyopia and strabismus for starters. Edwardian 07:22, 27 September 2005 (UTC)

[edit] Discussion3

I feel that the part that says "our depth perception was, in a way, created by trees", while being an amusing image, doesn't belong in a resource like wikipedia. Anyone agree/have evidence from the style guide?? 213.48.15.234 14:53, 20 February 2007 (UTC)

[edit] Monocular cues

Monocular cues are a particularly important subject for video games designers and quite complex ones are dirctly supported by high end video cards.

I believe Motion parallax is more important to humans than indicated. Nodding an object up and down on a screen with a frequency as would be seen when walking seems to give a much better impression of 3D than does waggling it side to side - intuitively one might have expected the side by side comparison to be better because of binoular vision - it'd be interesting to see if people who never could walk have the same experience. Dmcq 07:57, 24 June 2007 (UTC)

[edit] Philosophical implications

What the fuck is this shit? 121.45.15.32 15:28, 17 August 2007 (UTC)

That's right. I removed the section. See below:

Depth perception is (along with sexual reproduction), a convincing real-life example of the "thesis+antithesis>synthesis" model of progress developed by Hegel, Fichte, and Engels. Dialectics, including dialectical materialism, derives from the idea of a dialogue between people representing different points of view on a subject, who arrive through argument at a new way of seeing the subject that preserves whatever remains valid from both sides of the discussion (as in the Socratic dialogues of Plato). Binocular vision is a sort of argument between one eye (thesis) and the other (antithesis), each seeing the organism's environment from a slightly different perspective, which the brain resolves into a three-dimensional image containing contributions from both but transcending their limits. As in any good synthesis, depth perception is an almost magical leap to a higher level, embodying a "qualitative change" that could hardly have been imagined or predicted, by examining its component parts.

[edit] Squirrels

"and squirrels, which move in lines orthogonal to an object of interest to do the same"

Is there a more specific source for this? The supplied link is to the front page of purveslab.net, and I can't find anything there.

Koala man 00:14, 23 August 2007 (UTC)

[edit] binocular vision not necessary

"In modern terminology, stereopsis is depth perception from binocular vision through exploitation of parallax. Depth perception does indeed rely primarily on binocular vision, but it also uses many other monocular cues to form the final integrated perception. There are monocular cues that would be significant to a "one-eyed" person, and more complex inferred cues, that require both eyes to be perceiving stereo while the monocular cues are noted. This "third" group relies on processing within the brain of the person, as they see a full field of view with both eyes."

This article is not as bad as I expected -- but the lead is muddled. The plain fact of the matter is that depth perception does not require binocular vision. People who have binocular vision think it is much more essential than it seems to be to people who have never had it.-69.87.203.221 22:54, 11 September 2007 (UTC)

At sometime in the distant past, I saw a graph that came out of a US airforce study into depth perception mechanisms. They listed:

• Focus: The amount of muscle tension required to distort the lens of the eye to make a sharp image.
• Stereopsis: The amount by which the eyes have to point inwards in order to 'fuse' a single image.
• Parallax: Small head motions causing objects in the foreground to move by a bigger amount than those in the background.
• Object size: You know how big a car is in reality - so comparing that to the size on your retina tells you its distance.
• Atmospheric dispersion: At long distances, the attenuation of green light and the general mistiness of the scene can give you range cues.
• Earth curvature: At extremely long distances, the curvature of the earth itself becomes a distance cue.

Each mechanism was shown as being effective over some span of distances with a roughly gaussian shaped curve for each - but the brain is good at getting an accurate range estimate even when one or more of the mechanisms is giving false or inaccurate information. Very close up, you can't cross your eyes enough to fuse the image - so focus dominates range estimation out to a few feet. Out to (IIRC) 20' or so, stereopsis dominates - but beyond that, you can't measure the angles accurately enough and parallax takes over - and throughout the whole range of distances, object size and atmospheric dispersion have an effect. Only at very long ranges does earth curvature have an impact. If one or more of these mechanisms is impaired (eg if you are blind in one eye) then the other effects can compensate to some degree. Several observations back up these assertions:

• When people first start wearing glasses or contact lenses, they can find range estimation over very short distances difficult.
• Astronauts on the moon - robbed of object size by the fractal nature of cratering and having no atmospheric dispersion cues to fall back on found it hard to estimate distances beyond ~100 feet.
• A very common defect in vision at birth in about 10% of people results in a crossing of the eyes. If this is not attended to within a fairly short amount of time (weeks or months - but not years) then these people (with two perfectly good eyes) frequently fail to develop a stereopsis sense because even though the eye-crossing sorts itself out within a year, the part of the brain that deals with stereopsis never develops. In recent years, this is tested for - but generally, 5% of people do not use the stereopsis mechanism at all.

SteveBaker 15:47, 12 September 2007 (UTC)

Actually, stereopsis is not "The amount by which the eyes have to point inwards in order to 'fuse' a single image". I think you confused it for vergence. Stereopsis is normally referred to as the process of fusing the two different retinal images as a result of binocular disparity. --Arthur Lugtigheid 13:19, 27 October 2007 (UTC)

[edit] Proprioception plus Depth Perception?

The brain takes proprioception, depth perception, and other resources and uses them all together to create a 3-D understanding of the world surroundings and the physical self located in and moving in the world. What is the proper term for the whole functioning brain 3-D model/ability?-69.87.200.68 11:27, 12 September 2007 (UTC)

[edit] Brain vs. eyes

"In physiology, depth perception is the ability of an animal's brain to interpret visual information in three dimensions."

In some ways, the disambiguation article has the better short definition. Depth perception usually uses input from the eyes, but it is really a brain thing, using whatever information is available. (That's why movies and TV work so well.) -69.87.200.68 11:27, 12 September 2007 (UTC)

[edit] Untitled discussion

So what is the difference between one's performance on depth preception adn stereopsis? What dictates them? If one has different of 2 lines decimal visual acuity or more between eyes when uncorrected(and never had correction), would you expect a better depth perception or stereopsis after full correction to balance visual acuity? —Preceding unsigned comment added by 218.247.199.97 (talk) 12:57, 18 April 2008 (UTC)

Is this not just an aspect of visual perception? We need to decide whether that article contains all these elements, or if they are off in their own little articles, all cross-linked to each other and to other articles on related issues in the senses. For instance depth perception is related to motion and the kinesthetic sense since the relation of a moving body/observer to the observed environment is important to cognition and even in perception directly. All those attempts to get robots to come to conclusions about their environment with one eye and no neck, never worked. Two eyes didn't work. Only when the robots got two eyes and a neck, could they calculate their terrain with anything other than bizarre mistakes.

I think it deserves its own page, it is very important.

When somebody wants to integrate it. Just the basic facts, source Goldstein [(Goldstein, Sensation and Persception (Sixth Edition)].

[edit] The name of this article

This article's title is "Depth perception", but is is about depth perception only via vision. Yet we know that people can perceive the three-dimensionality of the world from senses other than vision. We can experience depth from stereophonic information from our two ears. We can close our eyes, and use our hands and bodies to feel our way around the world: haptics and kinesthesis. We could even block our eyes, ears, skin, and still tell when we are close to someone by smell, especially if he or she forgot to have a shower in the morning!

Can editors think of a more appropriate name for this article, or should these other sources of depth information be included?Robert P. O'Shea (talk) 16:57, 8 May 2008 (UTC)

Arguably, "Visual Depth Perception" would be a better title - but since most people understand "Depth Perception" to mean "Visual Depth Perception", we'd have to retain the present title as a redirect. If we ever get articles about other kinds of depth perception, then there would certainly be a case for a disambiguation page. Meanwhile, I don't see the need to change anything. SteveBaker (talk) 17:34, 19 May 2008 (UTC)