Non Max Suppression: Non-Max Supression (NMS) is a technique used to select one bounding box for an object if multiple bounding boxes were detected with varying probability scores by object detection algorithms(example: Faster R-CNN,YOLO).
NMS is implemented in this notebook using two methods
- Python using numpy in vectorized approach.
- Using tensorflow builtin function.
Requirements to run this notebook: 1)Python3 2) Numpy 3) Tensorflow