Given a list of coordinates of a coplanar plane $\left(pt_1, pt_2, pt_3, \cdots \right)$, how to compute the centroid of the coplanar plane?
One way to do it is to project the plane onto $XY$ and $YZ$ plane, but I don't really favor this approach as you have to check the orientation of the coplanar plane first before doing the projection and computing the centroid.
More specifically, I'm looking for a natural extension of the 2D centroid plane algorithm in 3D:
\begin{align}
C_x&=\frac1{6A}\sum_{i=0}^{n-1}(x_i+x_{i+1})(x_iy_{i+1}-x_{i+1}y_i)\\
C_y&=\frac1{6A}\sum_{i=0}^{n-1}(y_i+y_{i+1})(x_iy_{i+1}-x_{i+1}y_i)\\
A&=\frac12\sum_{i=0}^{n-1}(x_iy_{i+1}-x_{i+1}y_i)
\end{align}
Any idea?
Best Answer
You can take any two orthogonal vectors $\vec{e_1}$ and $\vec{e_2}$ on the plane and use them as a basis. You also need some point $(x_0, y_0, z_0)$ on the plane as origin.
Given point with coordinates $(x_1, y_1, z_1)$ on your plane you calculate it's coordinates with respect to new basis:
$x = (x_1 - x_0) e_{1x} + (y_1 - y_0) e_{1y} + (z_1 - z_0) e_{1z}$
$y = (x_1 - x_0) e_{2x} + (y_1 - y_0) e_{2y} + (z_1 - z_0) e_{2z}$
And after that you can apply your formulae to get $C_x$ and $C_y$. Those coordinates are easyly transformed back into original 3d coordinates:
$x = x_0 + e_{1x} C_x + e_{2x} C_y$
$y = y_0 + e_{1y} C_x + e_{2y} C_y$
$z = z_0 + e_{1z} C_x + e_{2z} C_y$