I’m wondering: For the Vex V5 (and IQ) AI Vision Sensor, on the VexCode configuration preview, the inner white square of the AprilTag is boxed. However, I am wondering if anyone (jpearman) knows if I am correct in thinking that the corners given to the robot are actually the outer black square of the tag. I feel that I may have figured this out a long time ago, but maybe not?
It should not be, but I know there were some tweaks for 1.1.6 to improve center coordinate detection. Same size and position both in vision utility and V5 below. If the tag is rotated, the width and height will be different as that defines a bounding box around the tag that matches how color and AI objects are described.