"Target Detection" Faster RCNN

One, Faster RCNN is a key step for target detection

1. Basic network for feature extraction;
2. Feature transmission RPN network for candidate frame extraction;
3. Layer classification classifies the candidate object frame, the candidate frame regression layer of (x, y, w, h)fine adjustment.

2. Faster RCNN network structure

1. Conv layers. As a CNN network target detection method, Faster RCNN first uses a set of basic conv+relu+pooling layers to extract image feature maps. The feature maps are shared for the subsequent RPN layer and fully connected layer.
2. Region Proposal Networks. The RPN network is used to generate region proposals. This layer uses softmax to determine whether the anchors are positive or negative, and then uses bounding box regression to correct the anchors to obtain accurate proposals.
3. RoI Pooling. This layer collects the input feature maps and proposals, extracts proposal feature maps after combining these information, and sends them to the subsequent fully connected layer to determine the target category.
4. Classification. Use proposal feature maps to calculate the category of the proposal, and again bounding box regression to obtain the final precise position of the detection frame.

三、Region Proposal Networks(RPN)

The RPN network is actually divided into two lines. The upper one uses softmax to classify anchors to obtain positive and negative classifications, and the lower one is used to calculate the bounding box regression offset for the anchors to obtain an accurate proposal. The final Proposal layer is responsible for synthesizing positive anchors and the corresponding bounding box regression offset to obtain proposals, and at the same time eliminate proposals that are too small and beyond the boundary. In fact, when the entire network reaches the Proposal Layer, it completes the function equivalent to target positioning.