My mesh has 4 vertices and 4 faces.

My mesh has 8 vertices and 12 faces.

Choice of colors:
Color1 [248/255, 248/255, 160/255]
Color2 [234/255, 141/255, 145/255]

R_relative = [[0, 1, 0], [-1, 0, 0], [0, 0, 1]]
T_relative = [0, 0, 0]

R_relative = [[1, 0, 0], [0, 1, 0], [0, 0, 1]]
T_relative = [0, 0, 2]

R_relative = [[1, 0, 0], [0, 1, 0], [0, 0, 1]]
T_relative = [0.5, -0.5, 0]

R_relative = [[0, 0, 1], [0, 1, 0], [-1, 0, 0]]
T_relative = [-3, 0, 3]

Discuss some of the tradeoffs between rendering as a mesh vs a point cloud:
- Rendering speed: It's faster to render a point cloud comparing to a mesh because it does not need to render "faces" but points only.
- Rendering quality: The rendering quality of a point cloud highly depends on the number of sample points. Since rendering a mesh includes surface information, it's fidelity is higher.
- Ease of use: In terms of reforming the model, it's easy to change a point cloud because it's just editing points, while it's harder to edit the faces information for a mesh. However, mesh has faces information and higher fidelity, so it's easy to use mesh for real-world use.
- Memory usage: A point cloud generally needs less memory usage than a mesh. But specifically, in a n-dimensional space, the space complexity of a point cloud would be P*n, where P is the number of points, while it is M**n, where M is the number of grid.

I download three meshes from Free3D, including a sofa, a TV, and a TV Stand. I re-texture each furniture with different patterns. The sofa has a floral texture, the TV stand has a wooden texture, and TV has uniform dark gray color. Also, I rescale each object to let the meshes have suitable sizes, and conduct mesh transformations to put them in the right position. The gif shows you the model from continously changing viewpoints.