Many examples in the repository still pass frame_resolution_wh as a parameter when using PolygonZone. Considering its deprecation, shall we update these examples for consistency? I'm happy to proceed with the changes if deemed appropriate.

Many examples in the repository still pass frame_resolution_wh as a parameter when using PolygonZone. Considering its deprecation, shall we update these examples for consistency? I'm happy to proceed with the changes if deemed appropriate.

Let's fix the test errors first.

Hi @onuralpszr,

I handle the initial test error but there are still failures in the pytest suite. After investigating the failures, I found that the frame_resolution_wh parameter, used for clipping bounding boxes in the trigger function, leads to errors as it expects the resolution of the video. The clipped_xyxy calculation within the trigger function clips the bounding boxes within the polygon, resulting in incorrect anchor point detection. This discrepancy causes test failures as shown in the traceback.

So, how should we proceed?

Hi @jeslinpjames and @onuralpszr, I just fixed the tests. We are ready to merge!

Many examples in the repository still pass frame_resolution_wh as a parameter when using PolygonZone. Considering its deprecation, shall we update these examples for consistency? I'm happy to proceed with the changes if deemed appropriate.

Hey @SkalskiP, should I put another PR with the examples changed?

Many examples in the repository still pass frame_resolution_wh as a parameter when using PolygonZone. Considering its deprecation, shall we update these examples for consistency? I'm happy to proceed with the changes if deemed appropriate.

Hey @SkalskiP, should I put another PR with the examples changed?

Please do so :)

But please test examples and their behavior as well. If dropping creates side effects, it would be nice to know it

But please test examples and their behavior as well. If dropping creates side effects, it would be nice to know it

Will do.

@jeslinpjames yup, please open second PR with changed examples. I fixed one bug when I fixed tests. So I don't expect behavior problems. But test anyway:)

@jeslinpjames yup, please open second PR with changed examples. I fixed one bug when I fixed tests. So I don't expect behavior problems. But test anyway:)

Hey, I tested out a couple of examples, and everything looks good. I'll put another PR soon.