Line Passing Through Two Points

Through two points $(x_0,y_0)$ and $(x_1, y_1)$ in Cartesian space $\mathbb{R}^2$ passes an implicit line with equation


\begin{displaymath}
(y_1 - y_0) x - (x_1 - x_0) y - y_1 x_0 + x_1 y_0 = 0
\end{displaymath} (1.63)

where it is clear that there are no singularities and all values are admissible. Letting $(d_x, d_y)$ denote the difference between the two points, the line passing through a point $(x_0,y_0)$ and directed along the vector $(d_x, d_y)$ has equation
\begin{displaymath}
d_y x - d_x y + y_0 d_x - x_0 d_y = 0
\end{displaymath} (1.64)

Generalizing to the n-dimensional case, the equation of the line in $\mathbb{R}^{n}$, passing through two points $\mathbf {p}$ and $\mathbf{q}$ written in homogeneous form, is the locus of points $\mathbf{x} \in \mathbb{R}^n$ such that

\begin{displaymath}
\mathbf{x} = (1-t) \mathbf{p} + t \mathbf{q} = \mathbf{p} + t ( \mathbf{q} - \mathbf{p} )
\end{displaymath} (1.65)

the equation of the parametric ray with $t \in \mathbb{R}$ scalar value. The values of $\mathbf {x}$ associated with values $t \in [0,1]$ are points internal to the segment $(\mathbf{p},\mathbf{q})$.

Using homogeneous coordinates, restricted to the two-dimensional Cartesian case, we obtain the following noteworthy result: the line with parameters $\mathbf{l}=(a,b,c)^{\top}$, passing through points $\mathbf{x}_1$ and $\mathbf{x}_2$, is obtained as

\begin{displaymath}
\mathbf{l}=\mathbf{x}_1 \times \mathbf{x}_2
\end{displaymath} (1.66)

since any point $\mathbf {x}$ must satisfy equation (1.62) to belong to the line.

Paolo medici
2026-10-01