Multizonality appears with an error. What am I doing wrong?

[epigraphe]

I did the recognition, it works great when the area is equal to the entire screen.

ZONE_POLYGON = np.array([
    [0, 0],
    [1, 0],
    [1, 1],
    [0, 1]
])

Now I need to make 4 zones inside the frame. I used this https://roboflow.github.io/polygonzone/ and got the data.

I put them instead of the first option, and I get an error. I tried the possible options, but I get an error.
Please, help.

ZONE_POLYGON = np.array([[[0.04, 0.03],[0.42, 0.03],[0.41, 0.51],[0.05, 0.51]],
[[0.49, 0.03],[0.96, 0.03],[0.96, 0.51],[0.5, 0.51]],
[[0.05, 0.61],[0.41, 0.6],[0.4, 0.94],[0.04, 0.94]],
[[0.5, 0.61],[0.95, 0.61],[0.94, 0.91],[0.49, 0.92]]])

def main():
    st.title("Camera")
    rtsp_url = 'rtsp://cam/realmonitor?channel=1&subtype=00&authbasic=[AUTH]'
    frame_width, frame_height = 2560, 1440

    cap = VideoCapture(rtsp_url)

    device = "0" if torch.cuda.is_available() else "cpu"
    if device == "0":
        torch.cuda.set_device(0)

    model = YOLO("yolov8x.pt")
    model.predict(source="0", show=False, stream=True, classes=0, conf=0.9)

    box_annotator = sv.BoxAnnotator(
        thickness=4,
        text_thickness=4,
        text_scale=2
    )

    zone_polygon = (ZONE_POLYGON * np.array([frame_width, frame_height])).astype(int)
    print("Zone Polygon Shape:", zone_polygon.shape)
    print("Zone Polygon Data:")
    print(zone_polygon)
    zone = sv.PolygonZone(polygon=zone_polygon, frame_resolution_wh=(frame_width, frame_height))
    zone_annotator = sv.PolygonZoneAnnotator(
        zone=zone,
        color=sv.Color.white(),
        thickness=6,
        text_thickness=6,
        text_scale=4
    )

Zone Polygon Shape: (4, 4, 2)
Zone Polygon Data:
[[[ 102   43]
  [1075   43]
  [1049  734]
  [ 128  734]]

 [[1254   43]
  [2457   43]
  [2457  734]
  [1280  734]]

 [[ 128  878]
  [1049  864]
  [1024 1353]
  [ 102 1353]]

 [[1280  878]
  [2432  878]
  [2406 1310]
  [1254 1324]]]

Uncaught app exception
Traceback (most recent call last):
  File "/usr/local/lib/python3.10/dist-packages/streamlit/runtime/scriptrunner/script_runner.py", line 541, in _run_script
    exec(code, module.__dict__)
  File "/home/ubuntu/AI/YOLO8/first.py", line 186, in <module>
    main()
  File "/home/ubuntu/AI/YOLO8/first.py", line 78, in main
    zone = sv.PolygonZone(polygon=zone_polygon, frame_resolution_wh=(frame_width, frame_height))
  File "/usr/local/lib/python3.10/dist-packages/supervision/detection/tools/polygon_zone.py", line 41, in __init__
    self.mask = polygon_to_mask(
  File "/usr/local/lib/python3.10/dist-packages/supervision/detection/utils.py", line 24, in polygon_to_mask
    cv2.fillPoly(mask, [polygon], color=1)
cv2.error: OpenCV(4.8.1) /io/opencv/modules/imgproc/src/drawing.cpp:2431: error: (-215:Assertion failed) p.checkVector(2, CV_32S) >= 0 in function 'fillPoly'

[SkalskiP]

Hi @epigraphe 👋🏻 The problem is that you simultaneously pass the coordinates of all PolygonZone's. You need to separate polygons and initiate each PolygonZone separately.

import numpy as np
import supervision as sv

ZONE_POLYGONS = np.array([
    [[0.04, 0.03], [0.42, 0.03], [0.41, 0.51], [0.05, 0.51]],
    [[0.49, 0.03], [0.96, 0.03], [0.96, 0.51], [0.5, 0.51]],
    [[0.05, 0.61], [0.41, 0.6], [0.4, 0.94], [0.04, 0.94]],
    [[0.5, 0.61], [0.95, 0.61], [0.94, 0.91], [0.49, 0.92]]
])

ZONE_POLYGONS = (ZONE_POLYGONS * np.array([width, frame_height])).astype(int)

zones = [
    sv.PolygonZone(
        polygon=polygon,
        frame_resolution_wh=(frame_width, frame_height)
    )
    for polygon in ZONE_POLYGONS
]

[epigraphe]

Thank you very much, everything worked.
You are a real genius)

[SkalskiP]

I'm glad I could help. ;)