BoxAreaConvention
Enums · AiDotNet.Enums
How a box’s width and height are measured from its (x1, y1, x2, y2) corners when computing areas and IoU.
For Beginners
Some detectors treat coordinates as pixel indices, so a box from pixel 0 to pixel 9 covers 10 pixels and its width is x2 - x1 + 1. Others treat coordinates as continuous positions, so the same box has width x2 - x1 = 9. The two give slightly different IoU values for the same boxes.
Fields
| Field | Summary |
|---|---|
BoxAreaConvention Continuous |
Corners are continuous coordinates: width = x2 - x1 and height = y2 - y1, as in ultralytics. |
BoxAreaConvention PixelInclusive |
Corners are inclusive pixel indices: width = x2 - x1 + 1 and height = y2 - y1 + 1, as in cython_bbox bbox_overlaps used by the official ByteTrack implementation. |
int value__ |