search

rafaelpadilla / review_object_detection_metrics

Object Detection Metrics. 14 object detection metrics: mean Average Precision (mAP), Average Recall (AR), Spatio-Temporal Tube Average Precision (STT-AP). This project supports different bounding box formats as in COCO, PASCAL, Imagenet, etc.
Other
1.08k stars 215 forks source link

Equation (9) and (11) in publication #114

Closed karl-gardner closed 1 year ago

karl-gardner commented 2 years ago

Hello Rafael,

I will be using an example like the one you have in the publication for my proposal defense, and I am trying to understand how you have derived equation (11) in the paper. I have gone through the math and do understand the following statement you make:

"Actually the N-point interpolation as defined by Equation (11) computes an AP value which is equal to the value computed by the Riemann integral in Equation (9) multiplied by (N−1)/N"

However, I don't understand why the AP value from the Riemann integral isn't used directly. Thus, I am wondering why the following value wouldn't be used:

image

Is this just from what people have used from in the previous implementations or maybe equation (8) in the paper?

Thanks,

Karl Gardner

github-actions[bot] commented 1 year ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

karl-gardner commented 1 year ago

Hello @rafaelpadilla,

Any thoughts?

Thanks,

Karl

github-actions[bot] commented 1 year ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.