1. Practicing with Cameras (15 Points)
1.1 360-degree Renders (5 points)
1.2 Re-creating the Dolly Zoom (10 points)
2. Practicing with Meshes (10 Points)
2.1 Constructing a Tetrahedron (5 points)
There are 4 vertices and 4 faces.
2.1 Constructing a Cube (5 points)
There are 8 vertices and 12 faces (2 coplaner triangular faces per side of cube).
3 Re-texturing a mesh (10 points)
Color1=[0.5, 0.0, 1.0] (purple), Color2=[1.0, 0.5, 0.0] (orange).
4. Camera Transformations (10 points)
Transform 1
R_relative=[
[0,1,0],
[-1,0,0],
[0,0,1],
],
T_relative=[0,0,0]
R_relative: Camera rotation 90 degrees about the
z-axis.
T_relative: No change.
Transform 2
R_relative=[
[0,0,1],
[0,1,0],
[-1,0,0],
],
T_relative=[-3,0,3]
R_relative: Camera rotation 90 degrees about the
y-axis.
T_relative: Move camera forward(right) 3 units in new
x-axis, 3 units backward in new z-axis to place rotated camera back in
front of cow.
Transform 3
R_relative=[
[1,0,0],
[0,1,0],
[0,0,1],
],
T_relative=[0,0,2]
R_relative: No change.
T_relative: Move camera 2 units forward in z-axis to
place camera further away from cow.
Transform 4
R_relative=[
[1,0,0],
[0,1,0],
[0,0,1],
],
T_relative=[0.5,-0.5,0]
R_relative: No change.
T_relative: Move camera 0.5 units forward(right) in
x-axis, and 0.5 units backward(up) in y-axis to translate camera off to
center.
5. Rendering Generic 3D Representations (45 Points)
5.1 Rendering Point Clouds from RGB-D Images (10 points)
5.2 Parametric Functions (10 + 5 points)
5.3 Implicit Surfaces (15 + 5 points)
Point clouds can be easier to use because the data encoded is
simpler. There is just a series of vertices in an unordered list
(N_v,3). The representation is invariant to permuations in
the vertices list, losing a few (or most) of the points. This allow’s
for flexibility. I can skip over rendering most of the points in the
point cloud if I want to do a light test. I don’t need to worry about
the order of the points. I just have one object in the code because all
my data sits in one unordered list.
Meshes can be more difficult to use because there is more information to be encoded. The vertices, like point clouds, are a list of points, but the order matters in this case. The faces may reference the vertices by index. The order of points within a face matters to the orientation of the face. The orientation of the face matters because the renderer often treats the outer and inner faces differently. The mesh is sensitive to each of these order changes, and there are more objects to keep track of and distinguish (vertices, faces, textures).
6. Do Something Fun (+10 points) (Extra Credit)
Parameterized cakes:
7. Sampling Points on Meshes (10 points) (Extra Credit)
Original mesh
Sampled pointclouds