added data layer for jittering, tiling layer for per pix softmax, tes…

…ts, example networks and visualizations. will remove legacy code later
brodyh · Dec 17, 2014 · c27a408 · c27a408
1 parent 9368c19
commit c27a408
Show file tree

Hide file tree

Showing 26 changed files with 3,634 additions and 99 deletions.
diff --git a/examples/filter_visualization_driving.ipynb b/examples/filter_visualization_driving.ipynb
diff --git a/examples/filter_visualization_driving_softmax.ipynb b/examples/filter_visualization_driving_softmax.ipynb
diff --git a/include/caffe/data_layers.hpp b/include/caffe/data_layers.hpp
@@ -97,6 +97,7 @@ class BasePrefetchingDataLayer :
   Blob<Dtype> prefetch_label_;
 };
 
+
 template <typename Dtype>
 class DataLayer : public BasePrefetchingDataLayer<Dtype> {
  public:
@@ -127,6 +128,69 @@ class DataLayer : public BasePrefetchingDataLayer<Dtype> {
   MDB_val mdb_key_, mdb_value_;
 };
 
+template <typename Dtype>
+class DrivingDataLayer :
+  public BaseDataLayer<Dtype>, public InternalThread {
+ public:
+  explicit DrivingDataLayer(const LayerParameter& param)
+      : BaseDataLayer<Dtype>(param) {}
+
+  virtual ~DrivingDataLayer();
+  virtual void DataLayerSetUp(
+      const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  void LayerSetUp(
+      const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+  virtual inline LayerParameter_LayerType type() const {
+    return LayerParameter_LayerType_DRIVING_DATA;
+  }
+
+  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+  virtual inline int ExactNumBottomBlobs() const { return 0; }
+  virtual inline int MinTopBlobs() const { return 1; }
+  virtual inline int MaxTopBlobs() const { return 1000; }
+
+  virtual void CreatePrefetchThread();
+  virtual void JoinPrefetchThread();
+
+ protected:
+  virtual void InternalThreadEntry();
+
+  // Assume network has tiling layer.
+  bool ReadBoundingBoxLabelToDatum(
+      const DrivingData& data, Datum* datum,
+      const int h_off, const int w_off);
+
+  // Use channels for the resolution of tiled convolution.
+  bool ReadBoundingBoxLabelToDatumLegacy(
+      const DrivingData& data, Datum* datum,
+      const int h_off, const int w_off);
+
+  unsigned int Rand();
+
+  vector<Blob<Dtype>*> prefetch_datas_;
+  vector<Blob<Dtype>*> prefetch_labels_;
+
+  shared_ptr<Caffe::RNG> rng_;
+
+  // LEVELDB
+  shared_ptr<leveldb::DB> db_;
+  shared_ptr<leveldb::Iterator> iter_;
+  // LMDB
+  MDB_env* mdb_env_;
+  MDB_dbi mdb_dbi_;
+  MDB_txn* mdb_txn_;
+  MDB_cursor* mdb_cursor_;
+  MDB_val mdb_key_, mdb_value_;
+};
+
+
 /**
  * @brief Provides data to the Net generated by a Filler.
  *
@@ -354,7 +418,8 @@ class MemoryDataLayer : public BaseDataLayer<Dtype> {
  public:
   explicit MemoryDataLayer(const LayerParameter& param)
       : BaseDataLayer<Dtype>(param), has_new_data_(false) {}
-  virtual void DataLayerSetUp(const vector<Blob<Dtype>*>& bottom,
+  virtual void DataLayerSetUp(
+      const vector<Blob<Dtype>*>& bottom,
       vector<Blob<Dtype>*>* top);
 
   virtual inline LayerParameter_LayerType type() const {

diff --git a/include/caffe/vision_layers.hpp b/include/caffe/vision_layers.hpp
@@ -350,6 +350,50 @@ class CuDNNPoolingLayer : public PoolingLayer<Dtype> {
 };
 #endif
 
+/**
+ * @brief for transforming a C x H x W blob into a
+ * C/t^2 x H*t x W*t blob where t is the tile dimension, and each pixel in
+ * the input is turned into a txt grid from the channels in row major order.
+ *
+ */
+template <typename Dtype>
+class TilingLayer : public Layer<Dtype> {
+ public:
+  explicit TilingLayer(const LayerParameter& param)
+      : Layer<Dtype>(param) {}
+  virtual void LayerSetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Reshape(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+  virtual inline LayerParameter_LayerType type() const {
+    return LayerParameter_LayerType_TILING;
+  }
+
+  virtual inline int ExactNumBottomBlobs() const { return 1; }
+  virtual inline int MinTopBlobs() const { return 1; }
+  // MAX POOL layers can output an extra top blob for the mask;
+  // others can only output the pooled inputs.
+  virtual inline int MaxTopBlobs() const { return 1; }
+
+ protected:
+  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const vector<bool>& propagate_down, vector<Blob<Dtype>*>* bottom);
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const vector<bool>& propagate_down, vector<Blob<Dtype>*>* bottom);
+
+  int input_channels_, output_channels_;
+  int input_width_, input_height_;
+  int output_width_, output_height_;
+  int count_per_input_map_, count_per_output_map_;
+  int tile_dim_, tile_dim_sq_;
+};
+
+
 }  // namespace caffe
 
 #endif  // CAFFE_VISION_LAYERS_HPP_