Components of modern object detection algorithms
The drawback of the R-CNN and Fast R-CNN techniques is that they have two disjointed networks – one to identify the regions that likely contain an object and the other to make corrections to the bounding box where an object is identified. Furthermore, both models require as many forward propagations as there are region proposals. Modern object detection algorithms focus heavily on training a single neural network and have the capability to detect all objects in one forward pass. The various components of a typical modern object detection algorithm are:
- Anchor boxes
- Region proposal network (RPN)
- Region of interest (RoI) pooling
Let’s discuss these in the following subsections (we’ll be focusing on anchor boxes and RPN as we discussed RoI pooling in the previous chapter).
Anchor boxes
So far, we have had region proposals coming from the selectivesearch method. Anchor boxes come...
