Non-Max Suppression is a post-processing method for object detection tasks.
In most cases, an object detection model will predict multiple boxes for a single object like the picture in my note. However, we don’t want this crowded output. We instead want only the relevant ones. Non-Max suppression helps find the right bounding boxes for us.
The algorithm follows the steps below.
1. Discard Boxes with low probability scores by setting a certain threshold.
2. Within the remaining bounding boxes, get the box with the highest score.
3. Discard any remaining boxes which have an IoU(Intersection Over Union) higher than the set threshold.
