Merge pull request #2705 from KonstantinMatskevich:face_recognition_labels_info
This commit is contained in:
commit
1f4ddbe5b6
@ -46,6 +46,15 @@ a unified access to all face recongition algorithms in OpenCV. ::
|
||||
|
||||
// Deserializes this object from a given cv::FileStorage.
|
||||
virtual void load(const FileStorage& fs) = 0;
|
||||
|
||||
// Sets additional information as pairs label - info.
|
||||
void setLabelsInfo(const std::map<int, string>& labelsInfo);
|
||||
|
||||
// Gets string information by label
|
||||
string getLabelInfo(const int &label);
|
||||
|
||||
// Gets labels by string
|
||||
vector<int> getLabelsByString(const string& str);
|
||||
};
|
||||
|
||||
|
||||
@ -70,6 +79,8 @@ Moreover every :ocv:class:`FaceRecognizer` supports the:
|
||||
|
||||
* **Loading/Saving** the model state from/to a given XML or YAML.
|
||||
|
||||
* **Setting/Getting labels info**, that is storaged as a string. String labels info is useful for keeping names of the recognized people.
|
||||
|
||||
.. note:: When using the FaceRecognizer interface in combination with Python, please stick to Python 2. Some underlying scripts like create_csv will not work in other versions, like Python 3.
|
||||
|
||||
Setting the Thresholds
|
||||
@ -293,6 +304,30 @@ to enable loading the model state. ``FaceRecognizer::load(FileStorage& fs)`` in
|
||||
turn gets called by ``FaceRecognizer::load(const string& filename)``, to ease
|
||||
saving a model.
|
||||
|
||||
FaceRecognizer::setLabelsInfo
|
||||
-----------------------------
|
||||
|
||||
Sets string information about labels into the model.
|
||||
.. ocv:function:: void FaceRecognizer::setLabelsInfo(const std::map<int, string>& labelsInfo)
|
||||
|
||||
Information about the label loads as a pair "label id - string info".
|
||||
|
||||
FaceRecognizer::getLabelInfo
|
||||
----------------------------
|
||||
|
||||
Gets string information by label.
|
||||
.. ocv:function:: string FaceRecognizer::getLabelInfo(const int &label)
|
||||
|
||||
If an unknown label id is provided or there is no label information assosiated with the specified label id the method returns an empty string.
|
||||
|
||||
FaceRecognizer::getLabelsByString
|
||||
---------------------------------
|
||||
Gets vector of labels by string.
|
||||
|
||||
.. ocv:function:: vector<int> FaceRecognizer::getLabelsByString(const string& str)
|
||||
|
||||
The function searches for the labels containing the specified substring in the associated string info.
|
||||
|
||||
createEigenFaceRecognizer
|
||||
-------------------------
|
||||
|
||||
|
@ -948,6 +948,14 @@ namespace cv
|
||||
// Deserializes this object from a given cv::FileStorage.
|
||||
virtual void load(const FileStorage& fs) = 0;
|
||||
|
||||
// Sets additional information as pairs label - info.
|
||||
void setLabelsInfo(const std::map<int, string>& labelsInfo);
|
||||
|
||||
// Gets string information by label
|
||||
string getLabelInfo(const int &label);
|
||||
|
||||
// Gets labels by string
|
||||
vector<int> getLabelsByString(const string& str);
|
||||
};
|
||||
|
||||
CV_EXPORTS_W Ptr<FaceRecognizer> createEigenFaceRecognizer(int num_components = 0, double threshold = DBL_MAX);
|
||||
|
@ -98,10 +98,84 @@ inline vector<_Tp> remove_dups(const vector<_Tp>& src) {
|
||||
return elems;
|
||||
}
|
||||
|
||||
// The FaceRecognizer2 class is introduced to keep the FaceRecognizer binary backward compatibility in 2.4
|
||||
// In master setLabelInfo/getLabelInfo/getLabelsByString should be virtual and _labelsInfo should be moved
|
||||
// to FaceRecognizer, that allows to avoid FaceRecognizer2 in master
|
||||
class FaceRecognizer2 : public FaceRecognizer
|
||||
{
|
||||
protected:
|
||||
// Stored pairs "label id - string info"
|
||||
std::map<int, string> _labelsInfo;
|
||||
|
||||
public:
|
||||
// Sets additional information as pairs label - info.
|
||||
virtual void setLabelsInfo(const std::map<int, string>& labelsInfo)
|
||||
{
|
||||
_labelsInfo = labelsInfo;
|
||||
}
|
||||
|
||||
// Gets string information by label
|
||||
virtual string getLabelInfo(int label) const
|
||||
{
|
||||
std::map<int, string>::const_iterator iter(_labelsInfo.find(label));
|
||||
return iter != _labelsInfo.end() ? iter->second : "";
|
||||
}
|
||||
|
||||
// Gets labels by string
|
||||
virtual vector<int> getLabelsByString(const string& str)
|
||||
{
|
||||
vector<int> labels;
|
||||
for(std::map<int,string>::const_iterator it = _labelsInfo.begin(); it != _labelsInfo.end(); it++)
|
||||
{
|
||||
size_t found = (it->second).find(str);
|
||||
if(found != string::npos)
|
||||
labels.push_back(it->first);
|
||||
}
|
||||
return labels;
|
||||
}
|
||||
|
||||
};
|
||||
|
||||
// Utility structure to load/save face label info (a pair of int and string) via FileStorage
|
||||
struct LabelInfo
|
||||
{
|
||||
LabelInfo():label(-1), value("") {}
|
||||
LabelInfo(int _label, const std::string &_value): label(_label), value(_value) {}
|
||||
int label;
|
||||
std::string value;
|
||||
void write(cv::FileStorage& fs) const
|
||||
{
|
||||
fs << "{" << "label" << label << "value" << value << "}";
|
||||
}
|
||||
void read(const cv::FileNode& node)
|
||||
{
|
||||
label = (int)node["label"];
|
||||
value = (std::string)node["value"];
|
||||
}
|
||||
std::ostream& operator<<(std::ostream& out)
|
||||
{
|
||||
out << "{ label = " << label << ", " << "value = " << value << "}";
|
||||
return out;
|
||||
}
|
||||
};
|
||||
|
||||
static void write(cv::FileStorage& fs, const std::string&, const LabelInfo& x)
|
||||
{
|
||||
x.write(fs);
|
||||
}
|
||||
|
||||
static void read(const cv::FileNode& node, LabelInfo& x, const LabelInfo& default_value = LabelInfo())
|
||||
{
|
||||
if(node.empty())
|
||||
x = default_value;
|
||||
else
|
||||
x.read(node);
|
||||
}
|
||||
|
||||
|
||||
// Turk, M., and Pentland, A. "Eigenfaces for recognition.". Journal of
|
||||
// Cognitive Neuroscience 3 (1991), 71–86.
|
||||
class Eigenfaces : public FaceRecognizer
|
||||
class Eigenfaces : public FaceRecognizer2
|
||||
{
|
||||
private:
|
||||
int _num_components;
|
||||
@ -154,7 +228,7 @@ public:
|
||||
// faces: Recognition using class specific linear projection.". IEEE
|
||||
// Transactions on Pattern Analysis and Machine Intelligence 19, 7 (1997),
|
||||
// 711–720.
|
||||
class Fisherfaces: public FaceRecognizer
|
||||
class Fisherfaces: public FaceRecognizer2
|
||||
{
|
||||
private:
|
||||
int _num_components;
|
||||
@ -211,7 +285,7 @@ public:
|
||||
// patterns: Application to face recognition." IEEE Transactions on Pattern
|
||||
// Analysis and Machine Intelligence, 28(12):2037-2041.
|
||||
//
|
||||
class LBPH : public FaceRecognizer
|
||||
class LBPH : public FaceRecognizer2
|
||||
{
|
||||
private:
|
||||
int _grid_x;
|
||||
@ -228,7 +302,6 @@ private:
|
||||
// old model data.
|
||||
void train(InputArrayOfArrays src, InputArray labels, bool preserveData);
|
||||
|
||||
|
||||
public:
|
||||
using FaceRecognizer::save;
|
||||
using FaceRecognizer::load;
|
||||
@ -327,6 +400,27 @@ void FaceRecognizer::load(const string& filename) {
|
||||
fs.release();
|
||||
}
|
||||
|
||||
void FaceRecognizer::setLabelsInfo(const std::map<int, string>& labelsInfo)
|
||||
{
|
||||
FaceRecognizer2* base = dynamic_cast<FaceRecognizer2*>(this);
|
||||
CV_Assert(base != 0);
|
||||
base->setLabelsInfo(labelsInfo);
|
||||
}
|
||||
|
||||
string FaceRecognizer::getLabelInfo(const int &label)
|
||||
{
|
||||
FaceRecognizer2* base = dynamic_cast<FaceRecognizer2*>(this);
|
||||
CV_Assert(base != 0);
|
||||
return base->getLabelInfo(label);
|
||||
}
|
||||
|
||||
vector<int> FaceRecognizer::getLabelsByString(const string& str)
|
||||
{
|
||||
FaceRecognizer2* base = dynamic_cast<FaceRecognizer2*>(this);
|
||||
CV_Assert(base != 0);
|
||||
return base->getLabelsByString(str);
|
||||
}
|
||||
|
||||
//------------------------------------------------------------------------------
|
||||
// Eigenfaces
|
||||
//------------------------------------------------------------------------------
|
||||
@ -423,6 +517,17 @@ void Eigenfaces::load(const FileStorage& fs) {
|
||||
// read sequences
|
||||
readFileNodeList(fs["projections"], _projections);
|
||||
fs["labels"] >> _labels;
|
||||
const FileNode& fn = fs["labelsInfo"];
|
||||
if (fn.type() == FileNode::SEQ)
|
||||
{
|
||||
_labelsInfo.clear();
|
||||
for (FileNodeIterator it = fn.begin(); it != fn.end();)
|
||||
{
|
||||
LabelInfo item;
|
||||
it >> item;
|
||||
_labelsInfo.insert(std::make_pair(item.label, item.value));
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
void Eigenfaces::save(FileStorage& fs) const {
|
||||
@ -434,6 +539,10 @@ void Eigenfaces::save(FileStorage& fs) const {
|
||||
// write sequences
|
||||
writeFileNodeList(fs, "projections", _projections);
|
||||
fs << "labels" << _labels;
|
||||
fs << "labelsInfo" << "[";
|
||||
for (std::map<int, string>::const_iterator it = _labelsInfo.begin(); it != _labelsInfo.end(); it++)
|
||||
fs << LabelInfo(it->first, it->second);
|
||||
fs << "]";
|
||||
}
|
||||
|
||||
//------------------------------------------------------------------------------
|
||||
@ -544,6 +653,17 @@ void Fisherfaces::load(const FileStorage& fs) {
|
||||
// read sequences
|
||||
readFileNodeList(fs["projections"], _projections);
|
||||
fs["labels"] >> _labels;
|
||||
const FileNode& fn = fs["labelsInfo"];
|
||||
if (fn.type() == FileNode::SEQ)
|
||||
{
|
||||
_labelsInfo.clear();
|
||||
for (FileNodeIterator it = fn.begin(); it != fn.end();)
|
||||
{
|
||||
LabelInfo item;
|
||||
it >> item;
|
||||
_labelsInfo.insert(std::make_pair(item.label, item.value));
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// See FaceRecognizer::save.
|
||||
@ -556,6 +676,10 @@ void Fisherfaces::save(FileStorage& fs) const {
|
||||
// write sequences
|
||||
writeFileNodeList(fs, "projections", _projections);
|
||||
fs << "labels" << _labels;
|
||||
fs << "labelsInfo" << "[";
|
||||
for (std::map<int, string>::const_iterator it = _labelsInfo.begin(); it != _labelsInfo.end(); it++)
|
||||
fs << LabelInfo(it->first, it->second);
|
||||
fs << "]";
|
||||
}
|
||||
|
||||
//------------------------------------------------------------------------------
|
||||
@ -743,6 +867,17 @@ void LBPH::load(const FileStorage& fs) {
|
||||
//read matrices
|
||||
readFileNodeList(fs["histograms"], _histograms);
|
||||
fs["labels"] >> _labels;
|
||||
const FileNode& fn = fs["labelsInfo"];
|
||||
if (fn.type() == FileNode::SEQ)
|
||||
{
|
||||
_labelsInfo.clear();
|
||||
for (FileNodeIterator it = fn.begin(); it != fn.end();)
|
||||
{
|
||||
LabelInfo item;
|
||||
it >> item;
|
||||
_labelsInfo.insert(std::make_pair(item.label, item.value));
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// See FaceRecognizer::save.
|
||||
@ -754,6 +889,10 @@ void LBPH::save(FileStorage& fs) const {
|
||||
// write matrices
|
||||
writeFileNodeList(fs, "histograms", _histograms);
|
||||
fs << "labels" << _labels;
|
||||
fs << "labelsInfo" << "[";
|
||||
for (std::map<int, string>::const_iterator it = _labelsInfo.begin(); it != _labelsInfo.end(); it++)
|
||||
fs << LabelInfo(it->first, it->second);
|
||||
fs << "]";
|
||||
}
|
||||
|
||||
void LBPH::train(InputArrayOfArrays _in_src, InputArray _in_labels) {
|
||||
@ -849,7 +988,6 @@ int LBPH::predict(InputArray _src) const {
|
||||
return label;
|
||||
}
|
||||
|
||||
|
||||
Ptr<FaceRecognizer> createEigenFaceRecognizer(int num_components, double threshold)
|
||||
{
|
||||
return new Eigenfaces(num_components, threshold);
|
||||
|
@ -39,20 +39,23 @@ static Mat toGrayscale(InputArray _src) {
|
||||
return dst;
|
||||
}
|
||||
|
||||
static void read_csv(const string& filename, vector<Mat>& images, vector<int>& labels, char separator = ';') {
|
||||
static void read_csv(const string& filename, vector<Mat>& images, vector<int>& labels, std::map<int, string>& labelsInfo, char separator = ';') {
|
||||
std::ifstream file(filename.c_str(), ifstream::in);
|
||||
if (!file) {
|
||||
string error_message = "No valid input file was given, please check the given filename.";
|
||||
CV_Error(CV_StsBadArg, error_message);
|
||||
}
|
||||
string line, path, classlabel;
|
||||
string line, path, classlabel, info;
|
||||
while (getline(file, line)) {
|
||||
stringstream liness(line);
|
||||
getline(liness, path, separator);
|
||||
getline(liness, classlabel);
|
||||
getline(liness, classlabel, separator);
|
||||
getline(liness, info, separator);
|
||||
if(!path.empty() && !classlabel.empty()) {
|
||||
images.push_back(imread(path, 0));
|
||||
labels.push_back(atoi(classlabel.c_str()));
|
||||
if(!info.empty())
|
||||
labelsInfo.insert(std::make_pair(labels.back(), info));
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -69,15 +72,17 @@ int main(int argc, const char *argv[]) {
|
||||
// These vectors hold the images and corresponding labels.
|
||||
vector<Mat> images;
|
||||
vector<int> labels;
|
||||
std::map<int, string> labelsInfo;
|
||||
// Read in the data. This can fail if no valid
|
||||
// input filename is given.
|
||||
try {
|
||||
read_csv(fn_csv, images, labels);
|
||||
read_csv(fn_csv, images, labels, labelsInfo);
|
||||
} catch (cv::Exception& e) {
|
||||
cerr << "Error opening file \"" << fn_csv << "\". Reason: " << e.msg << endl;
|
||||
// nothing more we can do
|
||||
exit(1);
|
||||
}
|
||||
|
||||
// Quit if there are not enough images for this demo.
|
||||
if(images.size() <= 1) {
|
||||
string error_message = "This demo needs at least 2 images to work. Please add more images to your data set!";
|
||||
@ -111,7 +116,9 @@ int main(int argc, const char *argv[]) {
|
||||
// cv::createEigenFaceRecognizer(10, 123.0);
|
||||
//
|
||||
Ptr<FaceRecognizer> model = createEigenFaceRecognizer();
|
||||
model->setLabelsInfo(labelsInfo);
|
||||
model->train(images, labels);
|
||||
|
||||
// The following line predicts the label of a given
|
||||
// test image:
|
||||
int predictedLabel = model->predict(testSample);
|
||||
@ -124,6 +131,8 @@ int main(int argc, const char *argv[]) {
|
||||
//
|
||||
string result_message = format("Predicted class = %d / Actual class = %d.", predictedLabel, testLabel);
|
||||
cout << result_message << endl;
|
||||
if( (predictedLabel == testLabel) && !model->getLabelInfo(predictedLabel).empty() )
|
||||
cout << format("%d-th label's info: %s", predictedLabel, model->getLabelInfo(predictedLabel).c_str()) << endl;
|
||||
// Sometimes you'll need to get/set internal model data,
|
||||
// which isn't exposed by the public cv::FaceRecognizer.
|
||||
// Since each cv::FaceRecognizer is derived from a
|
||||
|
Loading…
x
Reference in New Issue
Block a user