EGGN 512 - Lecture 5-2 3D-2D Transforms

EGGN 512 Computer Vision

[hishammoe]

I appreciate this lecture series a lot, I know this was done more than a decade ago but its value will remain as one of the greatest series in the computer vision community.

In the robot sensor to camera example, we already know the position of the object P with respect to the camera, which is [0, 4, 10], since this is already in camera coordinates we can say P(pix) = (KP(c))*1/z, which in fact gives us the correct answer.

It's simpler this way, and I assume you already know that and went the other way around just to let us know how extrinsic and intrinsic parameters interact together to move points from world coordinates to pixel location.