[][src]Crate sdfu

sdfu - Signed Distance Field Utilities

This is a small crate designed to help when working with signed distance fields in the context of computer graphics, especially ray-marching based renderers. Most of what is here is based on Inigo Quilez' excellent articles.

If you're using one of the more popular math libraries in Rust, then just enable the corresponding feature and hopefully all the necessary traits are already implemented for you so that you can just start passing in your Vec3s or whatever your lib calls them and you're off to the races! If not, then you can implement the necessary traits in the mathtypes module and still use this library with your own math lib.

This crate is built around the central trait SDF. This trait is structured in a similar way to how std::iter::Iterator works. Anything that implements SDF is able to return a distance from a point to its distance field. SDFs can be combined, modified, and otherwise used for various tasks by using the combinator methods on the SDF trait, or by directly using the structs that actually implement those combinators.

Most SDFs will be build up from one or more primitives being modified and combined together--the distance fields in the primitive module provide good starting points for this.


demo image

The image above was rendered with my own path tracing renderer, rayn, by leveraging sdfu. The SDF that is rendered above was created with the following code:

use sdfu::SDF;

let sdf = sdfu::Sphere::new(0.45)
        sdfu::Box::new(Vec3::new(0.25, 0.25, 1.5)))
        sdfu::Sphere::new(0.3).translate(Vec3::new(0.3, 0.3, 0.0)),
        sdfu::Sphere::new(0.3).translate(Vec3::new(-0.3, 0.3, 0.0)),
        sdfu::Box::new(Vec3::new(0.125, 0.125, 1.5)).translate(Vec3::new(-0.3, 0.3, 0.0)))
        sdfu::Box::new(Vec3::new(0.125, 0.125, 1.5)).translate(Vec3::new(0.3, 0.3, 0.0)))
        sdfu::Box::new(Vec3::new(1.5, 0.1, 0.1)).translate(Vec3::new(0.0, 0.3, 0.0)))
        sdfu::Box::new(Vec3::new(0.2, 2.0, 0.2)))
    .translate(Vec3::new(0.0, 0.0, -1.0));


pub use mathtypes::Dim2D;
pub use mathtypes::Dim3D;
pub use mathtypes::Dimension;
pub use primitives::*;



Modifiers for SDFs.


Operations you can perform to combine two SDFs.


A collection of primitive SDFs that may the modified using functions in mods or combined using functions in ops. Note that these primitives are always centered around the origin and that you must transform the point you are sampling into 'primitive-local' space. Functions are provided in mods to do this easier.


Other random utilities that are helpful when using SDFs in computer graphics applications, such as estimating normals.



The core trait of this crate; an implementor of this trait is able to take in a vector and return the min distance from that vector to a distance field.