larnt The 3D Line Art Engine

larnt is a vector-based 3D renderer written in Rust. It is used to produce 2D vector graphics (think SVGs) depicting 3D scenes.

The output of an OpenGL pipeline is a rastered image. The output of larnt is a set of 2D vector paths.

This project is a Rust rewrite of the original Go implementation by Michael Fogleman.

Examples

Click on the example image to jump to the code.

Installation

Add to your Cargo.toml:

[dependencies]
larnt = "0.1.0"

Features

• Primitives
  • Sphere
  • Cube
  • Triangle
  • Cylinder
  • Cone
  • 3D Functions
• Triangle Meshes
  • OBJ & STL
• Vector-based "Texturing"
• CSG (Constructive Solid Geometry) Operations
  • Intersection
  • Difference
• Output to PNG or SVG

How it Works

To understand how larnt works, it's useful to start with the Shape trait:

use larnt::{Box, Hit, Paths, Ray, Vector};

pub trait Shape {
    fn compile(&mut self) {}
    fn bounding_box(&self) -> Box;
    fn contains(&self, v: Vector, f: f64) -> bool;
    fn intersect(&self, r: Ray) -> Hit;
    fn paths(&self) -> Paths;
}

Each shape must provide some Paths which are 3D polylines on the surface of the solid. Ultimately anything drawn in the final image is based on these paths. These paths can be anything. For a sphere they could be lat/lng grid lines, a triangulated-looking surface, dots on the surface, etc. This is what we call vector-based texturing. Each built-in Shape ships with a default paths() function (e.g. a Cube simply draws the outline of a cube) but you can easily provide your own.

Each shape must also provide an intersect method that lets the engine test for ray-solid intersection. This is how the engine knows what is visible to the camera and what is hidden.

All of the Paths are chopped up to some granularity and each point is tested by shooting a ray toward the camera. If there is no intersection, that point is visible. If there is an intersection, it is hidden and will not be rendered.

The visible points are then transformed into 2D space using transformation matrices. The result can then be rendered as PNG or SVG.

The contains method is only needed for CSG (Constructive Solid Geometry) operations.

Hello World: A Single Cube

The Code

use larnt::{Cube, Scene, Vector};

fn main() {
    // create a scene and add a single cube
    let mut scene = Scene::new();
    scene.add(Cube::new(Vector::new(-1.0, -1.0, -1.0), Vector::new(1.0, 1.0, 1.0)));

    // define camera parameters
    let eye = Vector::new(4.0, 3.0, 2.0);    // camera position
    let center = Vector::new(0.0, 0.0, 0.0); // camera looks at
    let up = Vector::new(0.0, 0.0, 1.0);     // up direction

    // define rendering parameters
    let width = 1024.0;  // rendered width
    let height = 1024.0; // rendered height
    let fovy = 50.0;     // vertical field of view, degrees
    let znear = 0.1;     // near z plane
    let zfar = 10.0;     // far z plane
    let step = 0.01;     // how finely to chop the paths for visibility testing

    // compute 2D paths that depict the 3D scene
    let paths = scene.render(eye, center, up, width, height, fovy, znear, zfar, step);

    // render the paths in an image
    paths.write_to_png("out.png", width, height);

    // save the paths as an svg
    paths.write_to_svg("out.svg", width, height).expect("Failed to write SVG");
}

The Output

Custom Texturing

Suppose we want to draw cubes with vertical stripes on their sides, as shown in the skyscrapers example above. We can implement the Shape trait for a custom type.

use larnt::{Cube, Shape, Paths, Vector, Box, Hit, Ray};

struct StripedCube {
    cube: Cube,
    stripes: i32,
}

impl Shape for StripedCube {
    fn bounding_box(&self) -> Box {
        self.cube.bounding_box()
    }

    fn contains(&self, v: Vector, f: f64) -> bool {
        self.cube.contains(v, f)
    }

    fn intersect(&self, r: Ray) -> Hit {
        self.cube.intersect(r)
    }

    fn paths(&self) -> Paths {
        let mut paths = Vec::new();
        let (x1, y1, z1) = (self.cube.min.x, self.cube.min.y, self.cube.min.z);
        let (x2, y2, z2) = (self.cube.max.x, self.cube.max.y, self.cube.max.z);

        for i in 0..=self.stripes {
            let p = i as f64 / self.stripes as f64;
            let x = x1 + (x2 - x1) * p;
            let y = y1 + (y2 - y1) * p;
            paths.push(vec![Vector::new(x, y1, z1), Vector::new(x, y1, z2)]);
            paths.push(vec![Vector::new(x, y2, z1), Vector::new(x, y2, z2)]);
            paths.push(vec![Vector::new(x1, y, z1), Vector::new(x1, y, z2)]);
            paths.push(vec![Vector::new(x2, y, z1), Vector::new(x2, y, z2)]);
        }
        Paths::from_vec(paths)
    }
}

Now StripedCube instances can be added to the scene.

Constructive Solid Geometry (CSG)

You can easily construct complex solids using Intersection, Difference.

use larnt::{new_difference, new_intersection, radians, Cube, Cylinder, Matrix, Sphere, TransformedShape, Vector};
use std::sync::Arc;

let shape = new_difference(vec![
    new_intersection(vec![
        Arc::new(Sphere::new(Vector::default(), 1.0)),
        Arc::new(Cube::new(Vector::new(-0.8, -0.8, -0.8), Vector::new(0.8, 0.8, 0.8))),
    ]),
    Arc::new(Cylinder::new(0.4, -2.0, 2.0)),
    Arc::new(TransformedShape::new(
        Arc::new(Cylinder::new(0.4, -2.0, 2.0)),
        Matrix::rotate(Vector::new(1.0, 0.0, 0.0), radians(90.0)),
    )),
    Arc::new(TransformedShape::new(
        Arc::new(Cylinder::new(0.4, -2.0, 2.0)),
        Matrix::rotate(Vector::new(0.0, 1.0, 0.0), radians(90.0)),
    )),
]);

This is (Sphere & Cube) - (Cylinder | Cylinder | Cylinder).

Unfortunately, it's difficult to compute the joint formed at the boundaries of these combined shapes, so sufficient texturing is needed on the original solids for a decent result.