Skip to content

Latest commit

 

History

History
54 lines (52 loc) · 1.32 KB

03_matrices.md

File metadata and controls

54 lines (52 loc) · 1.32 KB

Matrices

Linear Transformation

  • All lines must remain lines
  • Origin must remain fixed in places
  • These two properties mean that grid lines stay parallel and evening spaced.
  • Only need to record where the basis vectors end in order to model the transformation
    • Since each vector began as a linear transformation of the basis vectors, the transformed vectors remain a linear combination of the new, transformed basis vectors $$\begin{bmatrix} a & c\ b & d \end{bmatrix}\begin{bmatrix} x\ y \end{bmatrix} = x \begin{bmatrix} a\ b \end{bmatrix} + y \begin{bmatrix} c\ d \end{bmatrix} = \begin{bmatrix} ax + by\ cx + dy \end{bmatrix}$$

Composition of Linear Transformations

  • Let $\begin{bmatrix} 0 & 1\ 0 & 1 \end{bmatrix}$ denote a shear matrix and $\begin{bmatrix} 0 & -1\ 1 & 0 \end{bmatrix}$ denote a rotation. Then the following is true: $$\begin{bmatrix} 1 & 1\ 0 & 1 \end{bmatrix} \bigg(\begin{bmatrix} 0 & -1\ 1 & 0 \end{bmatrix} \begin{bmatrix} x\ y \end{bmatrix} \bigg) = \begin{bmatrix} 1 & -1\ 1 & 0\ \end{bmatrix}\begin{bmatrix} x\ y \end{bmatrix}$$
    • These two operations are equivalent to applying the rotation and then the shear.
    • Read right to left, think function notation $f(g(x))$

Matrix Multiplication

  • $AB \ne BC$
  • $ABC = (AB)C = A(BC)$