“If you’ve been working with object detection long enough, you’ve undoubtedly encountered the problem of double detection.
For some reason, the model detects the same object multiple times on the same image. This is particularly inconvenient if you want to build more advanced analytics — like counting or tracking detections.
This exact issue occurred during a recent project, and it was quickly solved thanks to Non-Maximum Suppression (NMS). This post will explain how NMS works and shows you how to code NMS in NumPy.” - @SkalskiP