Talk:UV mapping

Computing Stub‑class

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing
Stub	This article has been rated as Stub-class on Wikipedia's content assessment scale.
???	This article has not yet received a rating on the project's importance scale.
	This article has been automatically rated by a bot or other tool as Stub-class because it uses a stub template. Please ensure the assessment is correct before removing the `\|auto=` parameter.

This needed an article --Shell 02:56, 19 December 2005 (UTC)[reply]

This article is awful

Sorry, but it just stinks. And the caption on the photo doesn't match the text. (One says "With and Without" and the other says "without and with") —Preceding unsigned comment added by 69.228.240.57 (talk) 06:05, August 26, 2007 (UTC)

Unfortunately, I'll have to agree. I came here hoping to learn what UV mapping is, but this article is in no way fit to explain it. I'll try to qualify:

First of all, I made the following assumptions based on my little knowledge of 3D modeling/rendering/mapping: Texture mapping basically means 'painting' a 3D object with an image, rather than with just a color or a pattern. Different geometric transformations might be used in the process, yielding different results.

UV mapping is a 3D modeling process of making a 2D image representing a 3D model.

Does "2D image" refer to the texture or a final rendered image? If it refers to the rendered image, that should be stated clearly in order to avoid confusion. If it refers to the texture, how is the modeling process involved with "making" the image?

The map transforms the 3D object onto an image known as a texture.

This is confusing. I somehow expected the image to be "transformed onto" the 3D object, not vice versa. But this may be what is going on 'behind the scenes', so I'm fine with it – just maybe add a couple of words to explain.

In contrast to "X", "Y" and "Z", which are the coordinates for the original 3D object in the modeling space, "U" and "V" are the coordinates of the transformed object.

Why mention X,Y,Z as though they were set in stone? It might as well be (r, θ, φ). What are we talking about here? Cartesian vs. Polar, 3D vs. 2D? X,Y,Z and U and V are just arbitrary names for variables. They explain nothing and have no inherit meaning.

This creates the effect of painting the image onto the surface of the 3D object.

That actually comes as quite a surprise at this point.

And then, a little later:

Without UV mapping, the checkers tile XYZ space and the sphere is carved out of them.

Huh? How can you carve a 3D object out of a 2D image?

So these are the 'impressions' of someone who doesn't know what UV-mapping is. Maybe it will help someone who does to improve the article. Thanks. --Ruper222 (talk) 02:13, 15 May 2008 (UTC)[reply]

Article needs expansion

Assuming UV mapping works the same way across most implementations, why not treat it as a kind of mathematical article and explain HOW it works, not just WHAT it is? --67.183.186.73 00:13, 23 August 2006 (UTC)[reply]

So which is which?

In XYZ, I know which coordinate means what in space. But what directions do U & V have? --24.249.108.133 17:58, 13 November 2006 (UTC)[reply]

UV are the coordinates in 2D (so the image you're wrapping around the 3D object) - you could see them as X and Y in 2D. - Simeon87 12:50, 15 November 2006 (UTC)[reply]

Yeah, also I believe X is red, Y is green, and Z is blue. KenFehling 05:12, 24 July 2007 (UTC)[reply]

"U" is "up-ness", and "V" is "velocity", that's what I've been able to gather from the article. So, on the surface of a 3d object, X Y and Z aren't really relevant because those 3 coordinates describe points in the whole 3d space - not limited to the 2d surface. It's like, sure, that surface exists in 3d space, the space of the surface itself is still more like 2d than it is like 3d. I guess it will just not be Euclidean. OK, so, "up-ness" tell you a point's distance in a direction arbitrarily defined in the 2d space, you know, you have to start with that, and once you defined the direction, and you have the distance in terms of "upness", then "velocity" becomes the other coordinate because this is (literally) a vector - (metaphorically) it means that it provides a complete coordinate space on the 2d surface. OK, at least, that's what I've been abe to figure out from the article. Understand? OK good I hope it helps. RGB is actually not related in any way, KenFehrling, but that was a good metap0hor too. Colorspace coordinate is almost as confusing in describing a particular color as UV coordinate is in describing the concept of surface-mapping. Cool? I didn't get it all myself until I read it OUT LOUD. It helps. --76.247.105.43 (talk) 03:07, 22 March 2009 (UTC)[reply]

Poor image example

The horrid image shows a triangle and a sphere that look no different than planar and spherical mapping. This needs a new image.

My thoughts exactly. --24.249.108.133 23:18, 28 February 2007 (UTC)[reply]

I created an image of UV mapping on a sphere, but it's still just spherical mapping. Some kind of mesh perhaps? —Josh Lee 20:37, 17 June 2007 (UTC)[reply]

terms not defined

$u=\sin \theta \cos \phi ={\frac {x}{\sqrt {x^{2}+y^{2}+z^{2}}}}$

What are theta and phi? --JWSchmidt 21:53, 9 July 2007 (UTC)[reply]

Poor example UV mapping

The example UV mapping given is a rather poor one, as it is just a projection of the sphere surface onto the x-y plane. This means that the top and bottom hemispheres will have the same UV coordinates. It would be much better to show an example UV mapping based on spherical polar coodinates (e.g. a latitude-longitude UV mapping). 121.73.55.195 (talk) 04:15, 5 June 2008 (UTC)[reply]

In fact the plane mapping formulas here don't correspond to the pictures at all - those show a true spherical mapping. Here are the formulae for spherical mapping from my own ray tracer (where $y$ is "up"): ${\hat {d}}=$ the normalised unit vector from the sphere's origin to the mapped point

$u=0.5-{\frac {\arctan 2(d_{z},d_{x})}{2\pi }}$
$v=0.5-2.0*{\frac {\arcsin(d_{y})}{2\pi }}$
Giving $u$ and $v$ both in the range $[0,1]$
Raybellis (talk) 09:12, 29 August 2010 (UTC)[reply]

Proposal new image

Would this image be valuable addition to this UV-Mapping page? At least this image relates more to the concepts of UV-mapping than the other pictures do so far.

Tschmits (talk) 09:49, 16 August 2008 (UTC)[reply]

Image correct?

Please correct me if I'm mistaken, but I don't think the 2d image represents the surface of the 3d sphere. In a sphere, all meridians will meet in the poles, right? Thus when folded in 3d, this image does not form a ball, but a cylinder. I think that when carved correctly, the 2d should resemble an orange peel - a series of lenslike shapes that add up to a sphere in 3d space. Cheers, Feirefis (talk) 22:52, 24 January 2010 (UTC)[reply]

You would think that, but the UV map does not have to conform to normal spacial relations. The map squares or tris, or any other poly, can be any size. A regualar poly sphere is made mostly of squares, when generated by the program, unless its a program that works primarily in triangles instead. But the UV map is shifted around and stretched out to be as even as possible, so all the faces are about the same, that way the image will fit onto each face without being stretched or distorted. It takes a lot of time to setup a UV map, and make sure all the faces match up and don't stretch out the image. So the end result usually consists of a squared up set of rectangles than what you would imagine an unfolded paper object to appear to look like - Aryeonos 71.93.34.210 (talk) 17:47, 25 August 2010 (UTC)[reply]