Octree is a data structure used for spatial partitioning, which can be used to efficiently locate nearby points in 3D space. This implementation of Octree is written in JavaScript and can be used in various applications that require spatial indexing of points.
https://octree-es.i.cnbbx.com/demo/main.html
https://octree-es.i.cnbbx.com/demo/demo.html
https://octree-es.i.cnbbx.com/demo/joinBoxGeometry.html
https://octree-es.i.cnbbx.com/demo/joinGltf.html
https://octree-es.i.cnbbx.com/demo/joinsphereGeometry.html
https://octree-es.i.cnbbx.com/demo/test.html
To create an Octree, you need to provide the position and size of the root cell. You can also specify the accuracy parameter, which determines the minimum distance between points in the octree. The default value of accuracy is 0.
import Octree from "./octree.js";
import Vec3 from "./vec3.js";
const origin = Vec3.create(0, 0, 0);
const size = Vec3.create(100, 100, 100);
const octree = new Octree(origin, size);
You can add points to the Octree using the add method. The second argument to this method is optional and can be used to associate some data with the point.
const point = Vec3.create(10, 20, 30);
const data = { name: "Point A" };
octree.add(point, data);
You can use the has method to check whether a point exists in the Octree or not. If a point exists, the method returns its associated data; otherwise, it returns false.
const point = Vec3.create(10, 20, 30);
if (octree.has(point)) {
console.log("The point exists in the octree");
} else {
console.log("The point does not exist in the octree");
}
You can also find the nearest point to a given point in the Octree using the findNearestPoint method. By default, this method returns only the nearest point. You can also specify the includeData option to include the associated data of the nearest point.
const point = Vec3.create(10, 20, 30);
const nearestPoint = octree.findNearestPoint(point);
if (nearestPoint) {
console.log("The nearest point is:", nearestPoint);
}
You can further restrict the search radius and exclude the given point using the maxDist and notSelf options.
const point = Vec3.create(10, 20, 30);
const options = { maxDist: 50, notSelf: true };
const nearestPoint = octree.findNearestPoint(point, options);
if (nearestPoint) {
console.log("The nearest point within 50 units is:", nearestPoint);
}
To find all points within a given radius, you can use the findNearbyPoints method.
const point = Vec3.create(10, 20, 30);
const radius = 50;
const results = octree.findNearbyPoints(point, radius);
if (results.points.length > 0) {
console.log(
"Found",
results.points.length,
"points within",
radius,
"units."
);
}
You can traverse all cells at a certain level using the getAllCellsAtLevel method. By default, this method starts with the root cell and returns all cells at the specified level that contain one or more points.
const level = 3;
const cells = octree.getAllCellsAtLevel(level);
console.log("Found", cells.length, "cells at level", level);
Octree is a simple yet powerful data structure for spatial indexing of points in 3D space. This implementation of Octree in JavaScript can be used in various applications that require fast and efficient lookup of nearby points.
The Vec3 class is a JavaScript implementation of a 3D vector that provides various operations for manipulating and calculating vectors in 3D space.
To create a new vector, you can use the constructor or the static create method:
import Vec3 from "./Vec3.js";
const v1 = new Vec3(1, 2, 3);
const v2 = Vec3.create(4, 5, 6);
The Vec3 class provides various operations for manipulating vectors. Here are some examples:
const v1 = Vec3.create(1, 2, 3);
const v2 = Vec3.create(4, 5, 6);
const result1 = v1.clone().add(v2); // result1: {x: 5, y: 7, z: 9}
const result2 = v1.clone().subtract(v2); // result2: {x: -3, y: -3, z: -3}
const v = Vec3.create(1, 2, 3);
const result1 = v.clone().multiplyScalar(2); // result1: {x: 2, y: 4, z: 6}
const result2 = v.clone().divideScalar(2); // result2: {x: 0.5, y: 1, z: 1.5}
const v1 = Vec3.create(1, 2, 3);
const v2 = Vec3.create(4, 5, 6);
const dotProduct = v1.dot(v2); // dotProduct: 32
const crossProduct = v1.clone().cross(v2); // crossProduct: {x: -3, y: 6, z: -3}
const v = Vec3.create(1, 2, 3);
const length = v.length(); // length: 3.7416573867739413
const normalizedVector = v.clone().normalize(); // normalizedVector: {x: 0.2672612419124244, y: 0.5345224838248488, z: 0.8017837257372732}
The Vec3 class also provides other operations such as set, copy, equals, negate, lerp, etc. You can refer to the source code for more information.
The Vec3 class provides a powerful and flexible way of working with 3D vectors in JavaScript. Its simple interface makes it easy to use in various applications that require vector calculations in 3D space.