Possibly wrong pose estimate in test

[tombelv]

If I'm interpreting the test_info.yaml file correctly, the position of the apriltag in the rotation_test should be at around 30 cm from the camera, while running the test and computing the euclidean distance from the resulting pose estimation returns around 20 cm.

The pose error estimate is in the order of 1e-6

Is there a calibration of configuration error? Or am I misintepreting the "ground truth"?

[afdaniele]

I'd ignore those tests.

[ajordan5]

I just ran the tests and noticed the same thing.

Could you provide context for why these tests should be ignored?

• Is ground truth incorrect in the yaml file?
• Is pose_t not the position of the tag relative to the camera frame?
• Possibly the K matrix is off for the camera used?

Outside of the provided tests I'm a bit lost on how to get started with implementing these bindings to estimate position and rotation of detected tags.

Thanks for any support you can provide!