Surfacing 101: Creating a UV Layout

Unwrapping a Model and Creating A UV Layout

Designer using laptop at office desk
Hero Images/ GettyImages

What is Surfacing?

By default, a recently finished 3D model is a lot like a blank canvas—most software packages will display it as an evenly lit, neutral shade of gray. No reflections, no color, no textures. Just plain old, boring gray.

Obviously, this isn't the way the model ultimately appears in the final render, so how is it that a model goes from an uninteresting shade of gray to the fully detailed characters and environments we see in movies and games?

Surfacing, which includes UV Layouts, texture mapping, and shader building, is the overall process of adding detail to the surface of a 3D object.

The job of a texturing or shader specialist may sound somewhat less glamorous than than that of a modeler or animator, but they're equally instrumental in the process of bringing a 3D film or game to fruition.

Try to imagine Rango without his colorful, scaly skin. Or Wall-E without his fantastically weathered-and-worn paint job. Without a good team of texture painters and shader writers any CG production will ultimately look flat and unconvincing.

Shading and texturing may be two sides of the same coin, but they're still fundamentally different processes, each deserving its own discussion. In this first section, we'll discuss UV layouts, and everything that goes along with creating them. In part-two we'll return with an explanation of texture-mapping, and then we'll round out the series with a quick look at shader networks.

Unwrapping a Model and Creating a UV Layout

Texture mapping, invented by Ed Catmull in 1974, is one of the more ingenious breakthroughs in the history of computer graphics. To put things in very general terms, texture mapping is the process of adding color (or other information) to a 3D model by projecting a two-dimensional image onto its surface.

However, in order to apply a texture map to the surface of a model, it first needs to be unwrapped and given a functional UV layout for texture artists to work with.

  • A UV layout is a visual representation of a 3D model flattened onto a two dimensional plane. Each point on the two-dimensional plane is called a UV and represents a vertex on the 3D object. In this way, all areas within the boundary of the UV layout correspond to a specific spot on the model. An easy way to visualize what a UV layout looks like is this:

    Have you ever made a cube out of paper? If so, think about the shape you needed to cut into the paper in order for it to properly fold into a cube—it probably looked like a cross, four units (faces) high and three across. If that paper cube were a 3D model, every fold would be an edge, every corner would be a vertex, and every flat area would be a face. Unwrapping a model is very similar, except there are usually more than six faces to flatten.
  • Unwrapping: Unwrapping is the process of assigning each polygonal face a set of UV coordinates on the two-dimensional image plane. UV coordinates are layed out visually and exported as a square bitmap image with a resolution anywhere from 512 x 512 all the way up to 6000 pixels by 6000. The artist who paints or creates the texture maps for the model will eventually use the layout for guidance when creating the texture files.
  • Creating a Functional UV Layout: The procedure for laying out a model's UV coordinates is usually three tiered:
     
    1. The artist will first select a group of faces on the model, and apply an automatic projection to those faces to provide a basis to work from. The projection is usually planar or cylindrical depending on the shape of the object:

      Planar Projection: A flat surface like a wall or floor would take a planar projection, which just means the 3D surface is flattened from a single direction (usually from the camera). The one caveat with planar projections is that it flattens all the way through a model—so if you applied a planar projection to an entire cube, most UVs would end up on top of one another.

      A cylindrical projection is better suited for curved surfaces, and works exactly as if you took a tube, made a cut from end to end, and unwrapped it until it was flat. There are other forms of projections as well, but the two discussed are the most common.
    2. Because most surfaces are not perfectly flat or cylindrical, automatic mapping techniques rarely give a satisfactory result. To combat this, the artist will manually tweak the UV layout by moving around UVs the way a modeler would push and pull vertices. A good UV layout aims to minimize seams, and dedicate a higher proportion of the allotted texture resolution to high detail areas of the mesh, like a character's face, or a particularly ornate part of their garment/armor.
    3. This process is repeated for all the polygons on the model. The artist will then go about arranging, or even merging UV groups to eliminate seams and create a logical, well organized UV layout.

    And that's it! Once the model is unwrapped, the process is placed into the hands of texture painters who will develop detailed image maps on top of the finished UV layout.