epl607_render

Team members

Stylanos Vassiliou

The programming language you chose

Python

The image processing library you chose for that language

PIL

Assignemnt 1

The process you used to render your test triangle

• To render my triangle, I used the rusterizaton process.
• I gave the program the 3 points of the triangle, its color, and the size of the image (coded in the program)
• It creates the bounding box by finding the minimum and maximum between the 3 points and the border, so that it will iterate only through the pixels within those bounds
• Then I calculate the a, b and c for each edge for the equation E(x, y) = ax + by + c
  • a is the rate of change on the edge for x
  • b is the rate of change on the edge for y
  • c is the point where the line intersects with the x-axis
• I then iterate over each pixel and check if it is within or at the edges, meaning that the E(x, y) = ax + by + c is positive for all 3, it colors it
  • by replacing the x,y with those of the pixel, i basicaly compare the position of the pixel relative of the line, to find if it is left, right, or on the line.
  • If it is right (or left) of all the lines, or on them, then it means that it is inside the triangle, or on it.

The test image you created

Assignemnt 2

Projection

• First, to project an image, we have to move the triangles so that the center of the image is the 0,0. By default, the 0,0 is the top left corner
• Then we divide x and y by z, to place the triangles in their correct projected position. If the depth is 0, we ignore the triangle

Rasterizarion

• The process is still, mostly the same as in Assignment 1
• After Projection a triangle, it finds its bounding boxes
• For each pixel:
  • It calculates the 3 edges and finds if a point os in the triangle
  • It gives a weight of how close a point is to the 3 vertexes of the triangle, to calculate the point's depth. This helps determine the color of triangles overlap, or clip
  • Finaly, if the depth is smaller than the current depth of the pixel, it colors it, and adds the depth to a depth buffer

Example image of overlapping triangles

Example image of clipping triangles

Example image of a pyramid

Final Assignemnt

Load 3D Geometry

• Downloaded object from The Stanford 3D Scanning Repository (https://graphics.stanford.edu/data/3Dscanrep/)
• Used blender to create a wavefront
• Used PyWavefront library to load the object
• When loading the object, the values were too small, so I scaled them up and placed them at WC 0, 0, 2
• Also when resizing, I made it so that the the smallest to biggest z had a difference of 2

Shading

• Used Phong equation to calculate local illumination.
• It simulates diffuce and speculat materials
• It uses a materia class that has 3 fields
  • diffuce: the diffuce coefficient, that has 3 values, 1 for each color of the RGB
  • specular: the specular coefficient that has 3 values too
  • shininess: Shininess controls how sharp and concentrated the specular highlight is on a surface.
• It also uses a lighting class
  • position: the position of the light
  • intensity: the intensity of the light for each color (3 values), from 0 to 1

Example of Diffuse material

• The light position is at 500, 500, -500

Example of Specular material

Camera

• The camera has the following fields
  • position: the xyz position of the camera
  • look at vector: the xyz vector for where to look
  • up vector: the xyz vector for where is up
  • field of view: how wide the camera will be in angles

Example of different cameras

• Front View

• Down View

• Up View

• Back View

• Left View

• Right View