Merge pull request #84 from avdmitry/datasetstools2

datasets loading module: continue

Author: vpisarev

modules/datasetstools/doc/datasetstools.rst

@@ -6,8 +6,6 @@ datasetstools. Tools for working with different datasets.
 
 The datasetstools module includes classes for working with different datasets.
 
-First version of this module was implemented for **Fall2014 OpenCV Challenge**.
-
 Action Recognition
 ------------------
 
@@ -50,13 +48,13 @@ FR_lfw
 
 Implements loading dataset:
 
-_`"Labeled Faces in the Wild-a"`: http://www.openu.ac.il/home/hassner/data/lfwa/
+_`"Labeled Faces in the Wild"`: http://vis-www.cs.umass.edu/lfw/
 
 .. note:: Usage
 
- 1. From link above download dataset file: lfwa.tar.gz.
+ 1. From link above download any dataset file: lfw.tgz\lfwa.tar.gz\lfw-deepfunneled.tgz\lfw-funneled.tgz and file with 10 test splits: pairs.txt.
 
- 2. Unpack it.
+ 2. Unpack dataset file and place pairs.txt in created folder.
 
  3. To load data run: ./opencv/build/bin/example_datasetstools_fr_lfw -p=/home/user/path_to_unpacked_folder/lfw2/
 
@@ -75,9 +73,11 @@ _`"ChaLearn Looking at People"`: http://gesture.chalearn.org/
 
  1. Follow instruction from site above, download files for dataset "Track 3: Gesture Recognition": Train1.zip-Train5.zip, Validation1.zip-Validation3.zip (Register on site: www.codalab.org and accept the terms and conditions of competition: https://www.codalab.org/competitions/991#learn_the_details There are three mirrors for downloading dataset files. When I downloaded data only mirror: "Universitat Oberta de Catalunya" works).
 
- 2. Unpack train archives Train1.zip-Train5.zip to one folder (currently loading validation files wasn't implemented)
+ 2. Unpack train archives Train1.zip-Train5.zip to folder Train/, validation archives Validation1.zip-Validation3.zip to folder Validation/
 
- 3. To load data run: ./opencv/build/bin/example_datasetstools_gr_chalearn -p=/home/user/path_to_unpacked_folder/
+ 3. Unpack all archives in Train/ & Validation/ in the folders with the same names, for example: Sample0001.zip to Sample0001/
+
+ 4. To load data run: ./opencv/build/bin/example_datasetstools_gr_chalearn -p=/home/user/path_to_unpacked_folders/
 
 GR_skig
 =======
@@ -239,13 +239,29 @@ Currently implemented loading full list with urls. Planned to implement dataset
 
  3. To load data run: ./opencv/build/bin/example_datasetstools_or_imagenet -p=/home/user/path_to_unpacked_file/
 
+OR_mnist
+===========
+.. ocv:class:: OR_mnist
+
+Implements loading dataset:
+
+_`"MNIST"`: http://yann.lecun.com/exdb/mnist/
+
+.. note:: Usage
+
+ 1. From link above download dataset files: t10k-images-idx3-ubyte.gz, t10k-labels-idx1-ubyte.gz, train-images-idx3-ubyte.gz, train-labels-idx1-ubyte.gz.
+
+ 2. Unpack them.
+
+ 3. To load data run: ./opencv/build/bin/example_datasetstools_or_mnist -p=/home/user/path_to_unpacked_files/
+
 OR_sun
 ======
 .. ocv:class:: OR_sun
 
 Implements loading dataset:
 
-_`"SUN Database"`: http://sun.cs.princeton.edu/
+_`"SUN Database"`: http://sundatabase.mit.edu/
 
 Currently implemented loading "Scene Recognition Benchmark. SUN397". Planned to implement also "Object Detection Benchmark. SUN2012". 
 

modules/datasetstools/include/opencv2/datasetstools/ar_hmdb.hpp

@@ -63,16 +63,9 @@ struct AR_hmdbObj : public Object
 class CV_EXPORTS AR_hmdb : public Dataset
 {
 public:
-    AR_hmdb() {}
-    AR_hmdb(const std::string &path, int number = 0);
-    virtual ~AR_hmdb() {}
+    virtual void load(const std::string &path) = 0;
 
-    virtual void load(const std::string &path, int number = 0);
-
-private:
-    void loadDataset(const std::string &path, int number = 0);
-
-    void loadAction(const std::string &fileName, std::vector<std::string> &train_, std::vector<std::string> &test_);
+    static Ptr<AR_hmdb> create();
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/ar_sports.hpp

@@ -63,16 +63,9 @@ struct AR_sportsObj : public Object
 class CV_EXPORTS AR_sports : public Dataset
 {
 public:
-    AR_sports() {}
-    AR_sports(const std::string &path);
-    virtual ~AR_sports() {}
+    virtual void load(const std::string &path) = 0;
 
-    virtual void load(const std::string &path, int number = 0);
-
-private:
-    void loadDataset(const std::string &path);
-
-    void loadDatasetPart(const std::string &fileName, std::vector< Ptr<Object> > &dataset_);
+    static Ptr<AR_sports> create();
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/dataset.hpp

@@ -62,10 +62,21 @@ class CV_EXPORTS Dataset
     Dataset() {}
     virtual ~Dataset() {}
 
-    virtual void load(const std::string &path, int number = 0) = 0;
+    virtual void load(const std::string &path) = 0;
 
-    std::vector< Ptr<Object> > train;
-    std::vector< Ptr<Object> > test;
+    std::vector< Ptr<Object> >& getTrain(int splitNum = 0);
+    std::vector< Ptr<Object> >& getTest(int splitNum = 0);
+    std::vector< Ptr<Object> >& getValidation(int splitNum = 0);
+
+    int getNumSplits() const;
+
+protected:
+    std::vector< std::vector< Ptr<Object> > > train;
+    std::vector< std::vector< Ptr<Object> > > test;
+    std::vector< std::vector< Ptr<Object> > > validation;
+
+private:
+    std::vector< Ptr<Object> > empty;
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/fr_lfw.hpp

@@ -56,21 +56,16 @@ namespace datasetstools
 
 struct FR_lfwObj : public Object
 {
-    std::string name;
-    std::vector<std::string> images;
+    std::string image1, image2;
+    bool same;
 };
 
 class CV_EXPORTS FR_lfw : public Dataset
 {
 public:
-    FR_lfw() {}
-    FR_lfw(const std::string &path);
-    virtual ~FR_lfw() {}
+    virtual void load(const std::string &path) = 0;
 
-    virtual void load(const std::string &path, int number = 0);
-
-private:
-    void loadDataset(const std::string &path);
+    static Ptr<FR_lfw> create();
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/gr_chalearn.hpp

@@ -80,14 +80,9 @@ struct GR_chalearnObj : public Object
 class CV_EXPORTS GR_chalearn : public Dataset
 {
 public:
-    GR_chalearn() {}
-    GR_chalearn(const std::string &path);
-    virtual ~GR_chalearn() {}
+    virtual void load(const std::string &path) = 0;
 
-    virtual void load(const std::string &path, int number = 0);
-
-private:
-    void loadDataset(const std::string &path);
+    static Ptr<GR_chalearn> create();
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/gr_skig.hpp

@@ -102,14 +102,9 @@ struct GR_skigObj : public Object
 class CV_EXPORTS GR_skig : public Dataset
 {
 public:
-    GR_skig() {}
-    GR_skig(const std::string &path);
-    virtual ~GR_skig() {}
+    virtual void load(const std::string &path) = 0;
 
-    virtual void load(const std::string &path, int number = 0);
-
-private:
-    void loadDataset(const std::string &path);
+    static Ptr<GR_skig> create();
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/hpe_parse.hpp

@@ -62,14 +62,9 @@ struct HPE_parseObj : public Object
 class CV_EXPORTS HPE_parse : public Dataset
 {
 public:
-    HPE_parse() {}
-    HPE_parse(const std::string &path);
-    virtual ~HPE_parse() {}
+    virtual void load(const std::string &path) = 0;
 
-    virtual void load(const std::string &path, int number = 0);
-
-private:
-    void loadDataset(const std::string &path);
+    static Ptr<HPE_parse> create();
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/ir_affine.hpp

@@ -64,14 +64,9 @@ struct IR_affineObj : public Object
 class CV_EXPORTS IR_affine : public Dataset
 {
 public:
-    IR_affine() {}
-    IR_affine(const std::string &path);
-    virtual ~IR_affine() {}
+    virtual void load(const std::string &path) = 0;
 
-    virtual void load(const std::string &path, int number = 0);
-
-private:
-    void loadDataset(const std::string &path);
+    static Ptr<IR_affine> create();
 };
 
 }

modules/datasetstools/include/opencv2/datasetstools/ir_robot.hpp

@@ -59,23 +59,23 @@ namespace datasetstools
 // 0.0000e+00   2.8285e+03   6.1618e+02
 // 0.0000e+00   0.0000e+00   1.0000e+00
 
+struct cameraPos
+{
+    std::vector<std::string> images;
+};
+
 struct IR_robotObj : public Object
 {
     std::string name;
-    std::vector<std::string> images; // TODO: implement more complex structure
+    std::vector<cameraPos> pos;
 };
 
 class CV_EXPORTS IR_robot : public Dataset
 {
 public:
-    IR_robot() {}
-    IR_robot(const std::string &path);
-    virtual ~IR_robot() {}
-
-    virtual void load(const std::string &path, int number = 0);
+    virtual void load(const std::string &path) = 0;
 
-private:
-    void loadDataset(const std::string &path);
+    static Ptr<IR_robot> create();
 };
 
 }