Crop layer for automatically aligning computations

include/caffe/common_layers.hpp

@@ -87,6 +87,9 @@ class ConcatLayer : public Layer<Dtype> {
  virtual inline const char* type() const { return "Concat"; }
  virtual inline int MinBottomBlobs() const { return 2; }
  virtual inline int ExactNumTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 
  protected:
  /**
@@ -165,6 +168,9 @@ class EltwiseLayer : public Layer<Dtype> {
  virtual inline const char* type() const { return "Eltwise"; }
  virtual inline int MinBottomBlobs() const { return 2; }
  virtual inline int ExactNumTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -289,6 +295,9 @@ class MVNLayer : public Layer<Dtype> {
  virtual inline const char* type() const { return "MVN"; }
  virtual inline int ExactNumBottomBlobs() const { return 1; }
  virtual inline int ExactNumTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -351,6 +360,9 @@ class SoftmaxLayer : public Layer<Dtype> {
  virtual inline const char* type() const { return "Softmax"; }
  virtual inline int ExactNumBottomBlobs() const { return 1; }
  virtual inline int ExactNumTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -414,6 +426,9 @@ class SplitLayer : public Layer<Dtype> {
  virtual inline const char* type() const { return "Split"; }
  virtual inline int ExactNumBottomBlobs() const { return 1; }
  virtual inline int MinTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -447,6 +462,9 @@ class SliceLayer : public Layer<Dtype> {
  virtual inline const char* type() const { return "Slice"; }
  virtual inline int ExactNumBottomBlobs() const { return 1; }
  virtual inline int MinTopBlobs() const { return 2; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,

include/caffe/layer.hpp

@@ -3,16 +3,20 @@
 
 #include <algorithm>
 #include <string>
+#include <utility>
 #include <vector>
 
 #include "caffe/blob.hpp"
 #include "caffe/common.hpp"
 #include "caffe/layer_factory.hpp"
 #include "caffe/proto/caffe.pb.h"
+#include "caffe/util/coords.hpp"
 #include "caffe/util/device_alternate.hpp"
 
 namespace caffe {
 
+template <typename Dtype> class Net;
+
 /**
  * @brief An interface for the units of computation which can be composed into a
  * Net.
@@ -285,6 +289,16 @@ class Layer {
  param_propagate_down_[param_id] = value;
  }
 
+ virtual DiagonalAffineMap<Dtype> coord_map() {
+ NOT_IMPLEMENTED;
+ // suppress warnings
+ return DiagonalAffineMap<Dtype>(vector<pair<Dtype, Dtype> >());
+ }
+
+ /**
+ * @brief Used by Net to give layers a pointer to their owning net.
+ */
+ void set_net(Net<Dtype>* net) { net_ = net; }
 
  protected:
  /** The protobuf that stores the layer parameters */
@@ -300,6 +314,9 @@ class Layer {
  * the objective function. */
  vector<Dtype> loss_;
 
+ /** The net to which this layer belongs. */
+ Net<Dtype>* net_;
+
  /** @brief Using the CPU device, compute the layer output. */
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
  const vector<Blob<Dtype>*>& top) = 0;

include/caffe/neuron_layers.hpp

@@ -32,6 +32,9 @@ class NeuronLayer : public Layer<Dtype> {
 
  virtual inline int ExactNumBottomBlobs() const { return 1; }
  virtual inline int ExactNumTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 };
 
 /**

include/caffe/util/coords.hpp

@@ -0,0 +1,61 @@
+#ifndef CAFFE_UTIL_COORDS_H_
+#define CAFFE_UTIL_COORDS_H_
+
+#include <algorithm>
+#include <utility>
+#include <vector>
+
+namespace caffe {
+
+template <typename Dtype>
+class DiagonalAffineMap {
+ public:
+ explicit DiagonalAffineMap(const vector<pair<Dtype, Dtype> > coefs)
+ : coefs_(coefs) { }
+ static DiagonalAffineMap identity(const int nd) {
+ return DiagonalAffineMap(vector<pair<Dtype, Dtype> >(nd, make_pair(1, 0)));
+ }
+
+ inline DiagonalAffineMap compose(const DiagonalAffineMap& other) const {
+ CHECK_EQ(coefs_.size(), other.coefs_.size())
+ << "Attempt to compose DiagonalAffineMaps of different dimensions";
+ DiagonalAffineMap<Dtype> out;
+ transform(coefs_.begin(), coefs_.end(), other.coefs_.begin(),
+ std::back_inserter(out.coefs_), &compose_coefs);
+ return out;
+ }
+ inline DiagonalAffineMap inv() const {
+ DiagonalAffineMap<Dtype> out;
+ transform(coefs_.begin(), coefs_.end(), std::back_inserter(out.coefs_),
+ &inv_coefs);
+ return out;
+ }
+ inline vector<pair<Dtype, Dtype> > coefs() { return coefs_; }
+
+ private:
+ DiagonalAffineMap() { }
+ static inline pair<Dtype, Dtype> compose_coefs(pair<Dtype, Dtype> left,
+ pair<Dtype, Dtype> right) {
+ return make_pair(left.first * right.first,
+ left.first * right.second + left.second);
+ }
+ static inline pair<Dtype, Dtype> inv_coefs(pair<Dtype, Dtype> coefs) {
+ return make_pair(1 / coefs.first, - coefs.second / coefs.first);
+ }
+ vector<pair<Dtype, Dtype> > coefs_;
+};
+
+template <typename Dtype>
+DiagonalAffineMap<Dtype> FilterMap(const int kernel_h, const int kernel_w,
+ const int stride_h, const int stride_w, const int pad_h, const int pad_w) {
+ vector<pair<Dtype, Dtype> > coefs;
+ coefs.push_back(make_pair(stride_h,
+ static_cast<Dtype>(kernel_h - 1) / 2 - pad_h));
+ coefs.push_back(make_pair(stride_w,
+ static_cast<Dtype>(kernel_w - 1) / 2 - pad_w));
+ return DiagonalAffineMap<Dtype>(coefs);
+}
+
+} // namespace caffe
+
+#endif // CAFFE_UTIL_COORDS_H_

include/caffe/vision_layers.hpp

@@ -162,6 +162,10 @@ class ConvolutionLayer : public BaseConvolutionLayer<Dtype> {
  : BaseConvolutionLayer<Dtype>(param) {}
 
  virtual inline const char* type() const { return "Convolution"; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return FilterMap<Dtype>(this->kernel_h_, this->kernel_w_, this->stride_h_,
+ this->stride_w_, this->pad_h_, this->pad_w_).inv();
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -195,8 +199,11 @@ class DeconvolutionLayer : public BaseConvolutionLayer<Dtype> {
  public:
  explicit DeconvolutionLayer(const LayerParameter& param)
  : BaseConvolutionLayer<Dtype>(param) {}
-
  virtual inline const char* type() const { return "Deconvolution"; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return FilterMap<Dtype>(this->kernel_h_, this->kernel_w_, this->stride_h_,
+ this->stride_w_, this->pad_h_, this->pad_w_);
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -314,6 +321,9 @@ class LRNLayer : public Layer<Dtype> {
  virtual inline const char* type() const { return "LRN"; }
  virtual inline int ExactNumBottomBlobs() const { return 1; }
  virtual inline int ExactNumTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return DiagonalAffineMap<Dtype>::identity(2);
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -396,6 +406,10 @@ class PoolingLayer : public Layer<Dtype> {
  return (this->layer_param_.pooling_param().pool() ==
  PoolingParameter_PoolMethod_MAX) ? 2 : 1;
  }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ return FilterMap<Dtype>(kernel_h_, kernel_w_, stride_h_, stride_w_,
+ pad_h_, pad_w_).inv();
+ }
 
  protected:
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
@@ -451,6 +465,39 @@ class CuDNNPoolingLayer : public PoolingLayer<Dtype> {
 };
 #endif
 
+template <typename Dtype>
+class CropLayer : public Layer<Dtype> {
+ public:
+ explicit CropLayer(const LayerParameter& param)
+ : Layer<Dtype>(param) {}
+ virtual void LayerSetUp(const vector<Blob<Dtype>*>& bottom,
+ const vector<Blob<Dtype>*>& top);
+ virtual void Reshape(const vector<Blob<Dtype>*>& bottom,
+ const vector<Blob<Dtype>*>& top);
+
+ virtual inline const char* type() const { return "Crop"; }
+ virtual inline int ExactNumBottomBlobs() const { return 2; }
+ virtual inline int ExactNumTopBlobs() const { return 1; }
+ virtual inline DiagonalAffineMap<Dtype> coord_map() {
+ vector<pair<Dtype, Dtype> > coefs;
+ coefs.push_back(make_pair(1, - crop_h_));
+ coefs.push_back(make_pair(1, - crop_w_));
+ return DiagonalAffineMap<Dtype>(coefs);
+ }
+
+ protected:
+ virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+ const vector<Blob<Dtype>*>& top);
+ virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+ const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom);
+ virtual void Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+ const vector<Blob<Dtype>*>& top);
+ virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+ const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom);
+
+ int crop_h_, crop_w_;
+};
+
 } // namespace caffe
 
 #endif // CAFFE_VISION_LAYERS_HPP_

src/caffe/layers/crop_layer.cpp

@@ -0,0 +1,126 @@
+#include <algorithm>
+#include <map>
+#include <set>
+#include <vector>
+
+#include "caffe/layer.hpp"
+#include "caffe/net.hpp"
+#include "caffe/vision_layers.hpp"
+
+namespace caffe {
+
+template <typename Dtype>
+void CropLayer<Dtype>::LayerSetUp(const vector<Blob<Dtype>*>& bottom,
+ const vector<Blob<Dtype>*>& top) {
+ // Construct a map from top blobs to layer inds, skipping over in-place
+ // connections.
+ map<Blob<Dtype>*, int> down_map;
+ for (int layer_ind = 0; layer_ind < this->net_->top_vecs().size();
+ ++layer_ind) {
+ vector<Blob<Dtype>*> tops = this->net_->top_vecs()[layer_ind];
+ for (int top_ind = 0; top_ind < tops.size(); ++top_ind) {
+ if (down_map.find(tops[top_ind]) == down_map.end()) {
+ down_map[tops[top_ind]] = layer_ind;
+ }
+ }
+ }
+ // Walk back from the first bottom, keeping track of all the blobs we pass.
+ set<Blob<Dtype>*> path_blobs;
+ Blob<Dtype>* blob = bottom[0];
+ int layer_ind;
+ // TODO this logic can be simplified if all blobs are tops
+ path_blobs.insert(blob);
+ while (down_map.find(blob) != down_map.end()) {
+ layer_ind = down_map[blob];
+ if (this->net_->bottom_vecs()[layer_ind].size() == 0) {
+ break;
+ }
+ blob = this->net_->bottom_vecs()[layer_ind][0];
+ path_blobs.insert(blob);
+ }
+ // Now walk back from the second bottom, until we find a blob of intersection.
+ Blob<Dtype>* inter_blob = bottom[1];
+ while (path_blobs.find(inter_blob) == path_blobs.end()) {
+ CHECK(down_map.find(inter_blob) != down_map.end())
+ << "Cannot align apparently disconnected blobs.";
+ layer_ind = down_map[inter_blob];
+ CHECK_GT(this->net_->bottom_vecs()[layer_ind].size(), 0)
+ << "Cannot align apparently disconnected blobs.";
+ inter_blob = this->net_->bottom_vecs()[layer_ind][0];
+ }
+ // Compute the coord map from the blob of intersection to each bottom.
+ vector<DiagonalAffineMap<Dtype> > coord_maps(2,
+ DiagonalAffineMap<Dtype>::identity(2));
+ for (int i = 0; i < 2; ++i) {
+ for (Blob<Dtype>* blob = bottom[i]; blob != inter_blob;
+ blob = this->net_->bottom_vecs()[down_map[blob]][0]) {
+ shared_ptr<Layer<Dtype> > layer = this->net_->layers()[down_map[blob]];
+ coord_maps[i] = coord_maps[i].compose(layer->coord_map());
+ }
+ }
+ // Compute the mapping from first bottom coordinates to second.
+ DiagonalAffineMap<Dtype> crop_map =
+ coord_maps[1].compose(coord_maps[0].inv());
+ for (int i = 0; i < 2; ++i) {
+ // Check for scale mismatch (unfortunately, CHECK_DOUBLE_EQ does not
+ // support a message like the other CHECKs).
+ CHECK_DOUBLE_EQ(crop_map.coefs()[i].first, 1);
+ CHECK_LE(crop_map.coefs()[i].second, 0) << "Negative crop width.";
+ // Check that the crop width is an integer.
+ CHECK_DOUBLE_EQ(crop_map.coefs()[i].second,
+ round(crop_map.coefs()[i].second));
+ }
+ crop_h_ = - round(crop_map.coefs()[0].second);
+ crop_w_ = - round(crop_map.coefs()[1].second);
+}
+
+template <typename Dtype>
+void CropLayer<Dtype>::Reshape(const vector<Blob<Dtype>*>& bottom,
+ const vector<Blob<Dtype>*>& top) {
+ top[0]->Reshape(bottom[0]->num(), bottom[0]->channels(), bottom[1]->height(),
+ bottom[1]->width());
+}
+
+template <typename Dtype>
+void CropLayer<Dtype>::Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+ const vector<Blob<Dtype>*>& top) {
+ const Dtype* bottom_data = bottom[0]->cpu_data();
+ Dtype* top_data = top[0]->mutable_cpu_data();
+ for (int n = 0; n < top[0]->num(); ++n) {
+ for (int c = 0; c < top[0]->channels(); ++c) {
+ for (int h = 0; h < top[0]->height(); ++h) {
+ caffe_copy(top[0]->width(),
+ bottom_data + bottom[0]->offset(n, c, crop_h_ + h, crop_w_),
+ top_data + top[0]->offset(n, c, h));
+ }
+ }
+ }
+}
+
+template <typename Dtype>
+void CropLayer<Dtype>::Backward_cpu(const vector<Blob<Dtype>*>& top,
+ const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom) {
+ const Dtype* top_diff = top[0]->cpu_diff();
+ Dtype* bottom_diff = bottom[0]->mutable_cpu_diff();
+ if (propagate_down[0]) {
+ caffe_set(bottom[0]->count(), static_cast<Dtype>(0), bottom_diff);
+ for (int n = 0; n < top[0]->num(); ++n) {
+ for (int c = 0; c < top[0]->channels(); ++c) {
+ for (int h = 0; h < top[0]->height(); ++h) {
+ caffe_copy(top[0]->width(),
+ top_diff + top[0]->offset(n, c, h),
+ bottom_diff + bottom[0]->offset(n, c, crop_h_ + h, crop_w_));
+ }
+ }
+ }
+ }
+}
+
+#ifdef CPU_ONLY
+STUB_GPU(CropLayer);
+#endif
+
+INSTANTIATE_CLASS(CropLayer);
+REGISTER_LAYER_CLASS(Crop);
+
+} // namespace caffe