ROIPoolingLayer源码解析
标签(空格分隔): 物体检测 faster-rcnn 源码
// ------------------------------------------------------------------
// Fast R-CNN
// Copyright (c) 2015 Microsoft
// Licensed under The MIT License [see fast-rcnn/LICENSE for details]
// Written by Ross Girshick
// ------------------------------------------------------------------
#include <cfloat>
#include "caffe/fast_rcnn_layers.hpp"
using std::max;
using std::min;
using std::floor;
using std::ceil;
/*
ROIPoolingLayer的作用将之前产生的Region Proposals pool成固定的大小的feature map
在faster-rcnn中有两组数据流向到ROIPoolingLayer
bottom[0]一组数据:是共享层的最后一层的feauter map 大小为(n,c,w,h);
bottom[1]另一组数据是:生成的regsion proposal,具体是batch的索引,以及对应的坐标 ,是一个二维数组大小为(number_anchor,5)
每一行的意思为(batch的索引,start_h,start_w,end_h,end_w) 注意坐标对应的原图像的坐标
ROPPoolingLayer输出为一系列的feature map 大小为(num_rois,channel_,pooled_w,pooled_h)
num_rois:产生Region Proposals的个数;
channel_: bottom[0]的通道数;
pooled_w,pooled_h:输出的feature的大小
*/
namespace caffe {
/*
LayerSetUp函数从prototxt文件中读取相关的参数
spatial_scale_:指定bottom[0]的feature map相对原图像的尺度
pooled_w,pooled_h:输出的feature的大小
*/
template <typename Dtype>
void ROIPoolingLayer<Dtype>::LayerSetUp(const vector<Blob<Dtype>*>& bottom,
const vector<Blob<Dtype>*>& top) {
ROIPoolingParameter roi_pool_param = this->layer_param_.roi_pooling_param();
CHECK_GT(roi_pool_param.pooled_h(), 0)
<< "pooled_h must be > 0";
CHECK_GT(roi_pool_param.pooled_w(), 0)
<< "pooled_w must be > 0";
pooled_height_ = roi_pool_param.pooled_h();
pooled_width_ = roi_pool_param.pooled_w();
spatial_scale_ = roi_pool_param.spatial_scale();
LOG(INFO) << "Spatial scale: " << spatial_scale_;
}
/*
Reshape函数确定输出数据top[0]的大小,所占的内存大小
(num_rois,channel_,pooled_w,pooled_h)
*/
template <typename Dtype>
void ROIPoolingLayer<Dtype>::Reshape(const vector<Blob<Dtype>*>& bottom,
const vector<Blob<Dtype>*>& top) {
channels_ = bottom[0]->channels();
height_ = bottom[0]->height();
width_ = bottom[0]->width();
top[0]->Reshape(bottom[1]->num(), channels_, pooled_height_,
pooled_width_);
max_idx_.Reshape(bottom[1]->num(), channels_, pooled_height_,
pooled_width_);
}
template <typename Dtype>
void ROIPoolingLayer<Dtype>::Forward_cpu(const vector<Blob<Dtype>*>& bottom,
const vector<Blob<Dtype>*>& top) {
const Dtype* bottom_data = bottom[0]->cpu_data();
const Dtype* bottom_rois = bottom[1]->cpu_data();
// Number of ROIs
int num_rois = bottom[1]->num(); //rois的个数
int batch_size = bottom[0]->num(); //batch_size
int top_count = top[0]->count();
Dtype* top_data = top[0]->mutable_cpu_data();
caffe_set(top_count, Dtype(-FLT_MAX), top_data);
int* argmax_data = max_idx_.mutable_cpu_data();
caffe_set(top_count, -1, argmax_data); //argmax_data赋予无限小的数据
// For each ROI R = [batch_index x1 y1 x2 y2]: max pool over R
for (int n = 0; n < num_rois; ++n) {
int roi_batch_ind = bottom_rois[0];//获取当前roi对应于batch的索引
//------将roi的坐标按照缩放比例映射到bottom[0]的尺度---------------//
int roi_start_w = round(bottom_rois[1] * spatial_scale_);
int roi_start_h = round(bottom_rois[2] * spatial_scale_);
int roi_end_w = round(bottom_rois[3] * spatial_scale_);
int roi_end_h = round(bottom_rois[4] * spatial_scale_);
//-----------------------------------------------------------//
CHECK_GE(roi_batch_ind, 0);
CHECK_LT(roi_batch_ind, batch_size);
int roi_height = max(roi_end_h - roi_start_h + 1, 1);
int roi_width = max(roi_end_w - roi_start_w + 1, 1);
//求取pooling的pooling kernel的窗口大小----------------//
const Dtype bin_size_h = static_cast<Dtype>(roi_height)
/ static_cast<Dtype>(pooled_height_);
const Dtype bin_size_w = static_cast<Dtype>(roi_width)
/ static_cast<Dtype>(pooled_width_);
//--------------------------------------------------------//
//获取对应batch的feature maps: 大小为(channel_,w,h)
const Dtype* batch_data = bottom_data + bottom[0]->offset(roi_batch_ind);
for (int c = 0; c < channels_; ++c) {
for (int ph = 0; ph < pooled_height_; ++ph) {
for (int pw = 0; pw < pooled_width_; ++pw) {
// Compute pooling region for this output unit:
// start (included) = floor(ph * roi_height / pooled_height_)
// end (excluded) = ceil((ph + 1) * roi_height / pooled_height_)
//根据roi放缩后的坐标找到在feature map上找到对应的位置(hstart,wstart,hend,wend)
int hstart = static_cast<int>(floor(static_cast<Dtype>(ph)
* bin_size_h));
int wstart = static_cast<int>(floor(static_cast<Dtype>(pw)
* bin_size_w));
int hend = static_cast<int>(ceil(static_cast<Dtype>(ph + 1)
* bin_size_h));
int wend = static_cast<int>(ceil(static_cast<Dtype>(pw + 1)
* bin_size_w));
//当然该矩形框必须要在feature map的范围之内,不能超过了feature map的大小
hstart = min(max(hstart + roi_start_h, 0), height_);
hend = min(max(hend + roi_start_h, 0), height_);
wstart = min(max(wstart + roi_start_w, 0), width_);
wend = min(max(wend + roi_start_w, 0), width_);
bool is_empty = (hend <= hstart) || (wend <= wstart);
//然后 找到 top[0]上feature map对应的点
const int pool_index = ph * pooled_width_ + pw;
if (is_empty) {
top_data[pool_index] = 0;
argmax_data[pool_index] = -1;
}
for (int h = hstart; h < hend; ++h) {
for (int w = wstart; w < wend; ++w) {
//在bottom[0]的featuer map上对应的点,该pooling是maxpool
const int index = h * width_ + w;
if (batch_data[index] > top_data[pool_index]) {
top_data[pool_index] = batch_data[index];
argmax_data[pool_index] = index;
}
}
}
}
}
// Increment all data pointers by one channel
batch_data += bottom[0]->offset(0, 1);
top_data += top[0]->offset(0, 1);
argmax_data += max_idx_.offset(0, 1);
}
// Increment ROI data pointer
bottom_rois += bottom[1]->offset(1);
}
}
template <typename Dtype>
void ROIPoolingLayer<Dtype>::Backward_cpu(const vector<Blob<Dtype>*>& top,
const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom) {
NOT_IMPLEMENTED;
}
#ifdef CPU_ONLY
STUB_GPU(ROIPoolingLayer);
#endif
INSTANTIATE_CLASS(ROIPoolingLayer);
REGISTER_LAYER_CLASS(ROIPooling);
} // namespace caffe