I’m reading a textbook on 3D graphics math. It’s describing how to model a pinhole camera, using a film plane at z = 1. However, the book then says this:
Given a point p in the scene with eye coordinates [x,y,z,1] it is easy enough to see (say, using a similar triangles argument) that the ray from p to the origin hits the plane at:
x_n = -(x/z)
y_n = -(y/z)
my question is, what do they mean by “using the similar triangles argument?” I’m aware similar triangles are ones with the same shape, but I don’t know what the “argument” they refer to here is, nor how it demonstrates the normals are negative [x or y] over z. I’m not very well versed in math, so apologies if this seems like a silly thing to not know.
Update: people asked for the associated diagram, here it is, or at least the closest one to the text above:

Perhaps the below annotated image with only the images that matter will help you understand better
This is a view of the triangle formed by point P on yz plane w.r.t pinhole (origin)
I considered a point on the yz plane, x = 0 to make it easier
Let P=[0,y,z] and it's reflection on the z=-1 plane be Q[0,y_n,-1]
You can verify that the triangles on the opposite sides of the y=0 plane in the above picture are similar triangles. So, from the rule of equal ratios of sides,
The similar triangles rule is used to derive the value of
y_nw.r.ty,zwhich are known coordinates. But the negativity comes from the fact that they are on the opposite sides of the y=0 planei.e. coordinate(y_n) = -(y/z)
You can do a similar setup with xz plane to derive x_n = -(x/z)