Quick Start.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PyTracer Quick Start Guide\n",
    "\n",
    "--------------\n",
    "\n",
    "\n",
    "\n",
    "In this quick start tutorial, we use a simple\n",
    "example to illustrate the basic work flow\n",
    "of PyTracer. Concretely, we will render\n",
    "a sphere (as a built-in `Sphere` object) and a\n",
    "slab (as modeled by a triangle mesh) under the\n",
    "illumination of a spherical diffuse area light.\n",
    "\n",
    "Essentially, we will model the scene and camera\n",
    "separately, hence the `Scene` object, after constructed.\n",
    "can be reused with different configuration\n",
    "of `Camera`. We shall thereafter initiate\n",
    "the main rendering loop and PyTracer will\n",
    "handle all the remaining work."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import (absolute_import, print_function, division)\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from pytracer import *\n",
    "# from pytracer import Spectrum\n",
    "\n",
    "# init spectrum (if applicable)\n",
    "# must be first initialized if\n",
    "# we use sampled texture/material,\n",
    "# e.g., sampled BRDF or SPD texture\n",
    "Spectrum.init()\n",
    "\n",
    "plt.axis('off')\n",
    "\n",
    "np.random.seed(1)\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Construction\n",
    "\n",
    "### Transformations\n",
    "\n",
    "\n",
    "`Transformation` objects are such that we use to specify\n",
    "how the entities are placed in thw world. PyTracer uses\n",
    "the left-handed coordinate system; each entity, including\n",
    "`Shape`s and `Camera`s are build in their object coordinate system.\n",
    "Different constructors\n",
    "of `Transform` can be used including `rotate`, `translate`, `look_at`\n",
    "and so on are defined for properly placing entities at convenience.\n",
    "\n",
    "We will manually specify the placement of each entity\n",
    "by way of `Transformation` in the sequel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from pytracer.geometry import (Vector, Point)\n",
    "from pytracer.transform import (Transform, AnimatedTransform)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Geometric Primitive and Light Construction\n",
    "\n",
    "We start by modeling a full sphere with radius $2$.\n",
    "Materials or textures also need to be specified for rendering.\n",
    "PyTracer has many build-in materials and textures. Generally,\n",
    "each `Material` is constructed with several `Texture` parameters\n",
    "specified. Each `Texture` can be constructed from a float scalar,\n",
    "or a `Spectrum` type.\n",
    "Measured `Texture` and `Material` are also supported, for example,\n",
    "the measured BRDF or SPD texture.\n",
    "\n",
    "Here we will use a simple `MatteMaterial` for the back slab; and\n",
    "a `UberMaterial` for the `Sphere`.\n",
    "\n",
    "Light source is also constructed at this stage, we build\n",
    "a dufuse area light source by attaching a `Sphere` object\n",
    "to it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from pytracer.shape import (Sphere, create_triangle_mesh)\n",
    "from pytracer.aggregate import GeometricPrimitive\n",
    "from pytracer.light import DiffuseAreaLight\n",
    "from pytracer.texture import ConstantTexture\n",
    "from pytracer.material import (MatteMaterial, UberMaterial)\n",
    "\n",
    "# slab\n",
    "back_param = {\n",
    "\t'indices': [0, 1, 2, 2, 0, 3],\n",
    "\t'P': [-20, 0, -10,\n",
    "          20, 0, -10,\n",
    "          20, 9, -10,\n",
    "          -20, 9, -10]\n",
    "}\n",
    "back_Kd = ConstantTexture(Spectrum([.1, .1, .1]))\n",
    "back_sigma = ConstantTexture(0.)\n",
    "back_mat = MatteMaterial(back_Kd, back_sigma)\n",
    "back_trans = Transform.translate(Vector(0., -5., 0.))\n",
    "back_shape = create_triangle_mesh(back_trans, back_trans.inverse(), False, back_param)\n",
    "back = GeometricPrimitive(back_shape, back_mat)\n",
    "\n",
    "\n",
    "# spheres\n",
    "sphere_transform = Transform.translate(Vector(.5, .5, -6))\n",
    "sphere_shape = Sphere(sphere_transform, sphere_transform.inverse(), False,  # whther reverse orientation\n",
    "                      2., -2., 2., 360.)  # for partial sphere: radius, z_min, z_max, phi_max (in degrees)\n",
    "\n",
    "# material\n",
    "Kd = ConstantTexture(Spectrum([.25, .25, .25]))\n",
    "Ks = ConstantTexture(Spectrum([.25, .25, .25]))\n",
    "Kr = ConstantTexture(Spectrum([0., 0., 0.]))\n",
    "Kt = ConstantTexture(Spectrum([.25, .25, .25]))\n",
    "roughness = ConstantTexture(0.1)\n",
    "opacity = ConstantTexture(Spectrum([1., 1., 1.]))\n",
    "eta = ConstantTexture(1.5)\n",
    "sphere_material = UberMaterial(Kd, Ks, Kr, Kt, roughness, opacity, eta)\n",
    "\n",
    "sphere = GeometricPrimitive(sphere_shape, sphere_material)\n",
    "shapes = [back, sphere]\n",
    "\n",
    "light_transform = Transform.translate(Vector(0., 1., 5.))\n",
    "light_shape = Sphere(light_transform, light_transform.inverse(), False, 1., -1., 1., 360.)\n",
    "light = DiffuseAreaLight(light_transform, Spectrum(10.), 2, light_shape)\n",
    "\n",
    "lights = [light]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Aggregation and Acceleration Structure\n",
    "\n",
    "Now we will make several encapsulations.\n",
    "First we pack all `Shape`s in the scene into some acceleration structures,\n",
    "here we select Bounding Volume Hierarchy (`BVH`) for fast ray-shape intersection\n",
    "test, though we only have one two `Shape`s; next we encapsulate the aggregation\n",
    "and lights to a `Scene` object. `Scene` objects are later passed to\n",
    "some `Renderer` for rendereing. It is designed to be reusable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from pytracer.scene import Scene\n",
    "from pytracer.aggregate import BVH\n",
    "aggs = BVH(shapes, False)\n",
    "scene = Scene(aggs, lights, None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Camera Construction\n",
    "\n",
    "### Camera and Sub-pixel Sampling\n",
    "\n",
    "\n",
    "Next we will configure the `Camera` placement.\n",
    "`Camera` are assumed to be movable, with movements interpolated\n",
    "using quaternion interpolation; this functionality has not\n",
    "been fully implemented though.\n",
    "PyTracer supports projective, perspective and\n",
    "orthographic cameras. Here we use a `ProjectiveCamera`.\n",
    "Noted for multiplcation is overloaded for `Transform` for\n",
    "cascading transformations.\n",
    "\n",
    "Each `Camera` possesses a `Film` object responsible\n",
    "for pixel updating and writing traced image to file. PyTracer\n",
    "uses sub-pixel sampling, hence a `Filter` is needed for `Film`\n",
    "to properly handle the interpolation.\n",
    "Sub-pixel samples are generated by the `Sampler`, which internally\n",
    "uses the `numpy` random generator."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from pytracer.camera import PerspectiveCamera\n",
    "from pytracer.film import ImageFilm\n",
    "from pytracer.filter import LanczosSincFilter\n",
    "from pytracer.sampler import StratifiedSampler\n",
    "\n",
    "\n",
    "# image resolution\n",
    "x_res, y_res = 64, 64\n",
    "\n",
    "# samples per pixel\n",
    "x_spr, y_spr = 2, 2\n",
    "\n",
    "# specify sampler\n",
    "sampler = StratifiedSampler(0, x_res, 0, y_res, x_spr, y_spr, True, 0., 0.)\n",
    "\n",
    "# specify placement for camera\n",
    "trans = Transform.translate(Vector(3.5, 3.5, 1.)) * Transform.look_at(Point(0.1, 0., 0.), Point(0., 0., -1.), Vector(0., 0., 1.))\n",
    "cam_trans = AnimatedTransform(trans, 0., trans, 0.)\n",
    "\n",
    "# create film\n",
    "# (xr, yr) is the image resolution in pixels\n",
    "# if no file name specified, a temporary file\n",
    "# `.tmp.png` is created\n",
    "film = ImageFilm(xr=x_res, yr=y_res, filt=LanczosSincFilter(4., 4., 3.),\n",
    "                 crop=[0., 1., 0., 1.], fn='tmp.png')\n",
    "\n",
    "camera = PerspectiveCamera(cam_trans, scr_win=[0., 1., 0., 1.], s_open=0.,\n",
    "                           s_close=0., lensr=0., focald=1e100, fov=90., f=film)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Light Transport and Rendering Loop\n",
    "\n",
    "Finally, we specify the sampler, integrators\n",
    "and renderer being used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from pytracer.integrator import DirectLightingIntegrator\n",
    "from pytracer.renderer import SamplerRenderer\n",
    "\n",
    "surface_integrator = DirectLightingIntegrator()\n",
    "renderer = SamplerRenderer(sampler, camera, surface_integrator, None)  # VolumeIntegrator: coming soon"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hereafter we may enter the main rendering loop.\n",
    "A progress bar is displayed for reporting the\n",
    "current status. Images are *always* written on\n",
    "completion, by default to `tmp.png`. A `numpy` array\n",
    "of shape `x_res * y_res * 3` is also returned. Internally,\n",
    "we use Pillow for image I/O."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rendering |████████████████████████████████████████████████████████████████████████████████████████████████████| 100.0% \n"
     ]
    }
   ],
   "source": [
    "image = renderer.render(scene)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x10e54c0b8>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPwAAAD7CAYAAABOrvnfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmvBCFIQmvBAFoQkvREFowgtREP8PWdurcPHzQhEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10e2a19e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Models\n",
    "\n",
    "We may also direct `import` constructed models.\n",
    "The saperation between `Scene` and `Camera` in\n",
    "rendering enables reusing the `Scene` object\n",
    "once loaded. We will next load the head mesh\n",
    "model and trace it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rendering |████████████████████████████████████████████████████████████████████████████████████████████████████| 100.0% \n"
     ]
    }
   ],
   "source": [
    "import examples.head_mesh as model\n",
    "\n",
    "# construct camera with preferred configuration\n",
    "camera = model.configure_camera(x_res=128, y_res=128)\n",
    "sampler = StratifiedSampler(0, 128, 0, 128, 1, 1, True, 0., 0.)\n",
    "surf_int = DirectLightingIntegrator()\n",
    "\n",
    "renderer = SamplerRenderer(sampler, camera, surf_int, None)\n",
    "image = renderer.render(model.scene)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x10f04ca90>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQEAAAD7CAYAAABqkiE2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10e5f2f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(image)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}