Faster RCNN 添加 ROIPoolingLayer 和SmoothL1LossLayer

想要在现有的caffe版本添加ROIPoolingLayer 和SmoothL1LossLayer，首先到这里https://github.com/rbgirshick/caffe-fast-rcnn可以找到他们的定义。caffe添加layer的详细方法在这里可以看到https://github.com/BVLC/caffe/wiki/，然后我们就可以开始啦！！

1.ROIPoolingLayer 和SmoothL1LossLayer的声明与实现：

声明：caffe_root/include/caffe/fast_rcnn_layers.hpp

实现：caffe_root/src/caffe/layers/roi_pooling_layer.cpp

    caffe_root/src/caffe/layers/roi_pooling_layer.cu

    caffe_root/src/caffe/layers/smooth_L1_loss_layer.cpp

    caffe_root/src/caffe/layers/smooth_L1_loss_layer.cu

2.ROIPoolingLayer 和SmoothL1LossLayer的测试：

caffe_root/src/caffe/test/test_roi_pooling_layer.cpp

caffe_root/src/caffe/test/test_smooth_L1_loss_layer.cpp

3.注册

caffe_root/src/caffe/proto/caffe.proto添加

optional ROIPoolingParameter roi_pooling_param = 8266711;

// Message that stores parameters used by ROIPoolingLayer
message ROIPoolingParameter {
// Pad, kernel size, and stride are all given as a single value for equal
// dimensions in height and width or as Y, X pairs.
  optional uint32 pooled_h = 1 [default = 0]; // The pooled output height
  optional uint32 pooled_w = 2 [default = 0]; // The pooled output width
  // Multiplicative spatial scale factor to translate ROI coords from their
  // input scale to the scale used when pooling
  optional float spatial_scale = 3 [default = 1];
}

optional SmoothL1LossParameter smooth_l1_loss_param = 8266712;

message SmoothL1LossParameter {
  // SmoothL1Loss(x) =
  //   0.5 * (sigma * x) ** 2    -- if x < 1.0 / sigma / sigma
  //   |x| - 0.5 / sigma / sigma -- otherwise
  optional float sigma = 1 [default = 1];
}

然后就大功告成了，重新编译caffe吧。