diff --git a/openbr/plugins/quality.cpp b/openbr/plugins/quality.cpp
index 469009e..11d4bbd 100644
--- a/openbr/plugins/quality.cpp
+++ b/openbr/plugins/quality.cpp
@@ -77,6 +77,12 @@ class ImpostorUniquenessMeasureTransform : public Transform
 
 BR_REGISTER(Transform, ImpostorUniquenessMeasureTransform)
 
+
+float KDEPointer(const QList<float> *scores, double x, double h)
+{
+    return Common::KernelDensityEstimation(*scores, x, h);
+}
+
 /* Kernel Density Estimator */
 struct KDE
 {
@@ -85,20 +91,35 @@ struct KDE
     QList<float> bins;
 
     KDE() : min(0), max(1), mean(0), stddev(1) {}
-    KDE(const QList<float> &scores)
+
+    KDE(const QList<float> &scores, bool trainKDE)
     {
         Common::MinMax(scores, &min, &max);
         Common::MeanStdDev(scores, &mean, &stddev);
+
+        if (!trainKDE)
+            return;
+
         double h = Common::KernelDensityBandwidth(scores);
         const int size = 255;
         bins.reserve(size);
-        for (int i=0; i<size; i++)
-            bins.append(Common::KernelDensityEstimation(scores, min + (max-min)*i/(size-1), h));
+
+        QFutureSynchronizer<float> futures;
+
+        for (int i=0; i < size; i++)
+            futures.addFuture(QtConcurrent::run(KDEPointer, &scores, min + (max-min)*i/(size-1), h));
+        futures.waitForFinished();
+
+        foreach(const QFuture<float> & future, futures.futures())
+            bins.append(future.result());
     }
 
     float operator()(float score, bool gaussian = true) const
     {
         if (gaussian) return 1/(stddev*sqrt(2*CV_PI))*exp(-0.5*pow((score-mean)/stddev, 2));
+        if (bins.empty())
+            return -std::numeric_limits<float>::max();
+	
         if (score <= min) return bins.first();
         if (score >= max) return bins.last();
         const float x = (score-min)/(max-min)*bins.size();
@@ -123,8 +144,8 @@ struct MP
 {
     KDE genuine, impostor;
     MP() {}
-    MP(const QList<float> &genuineScores, const QList<float> &impostorScores)
-        : genuine(genuineScores), impostor(impostorScores) {}
+    MP(const QList<float> &genuineScores, const QList<float> &impostorScores, bool trainKDE)
+        : genuine(genuineScores, trainKDE), impostor(impostorScores, trainKDE) {}
     float operator()(float score, bool gaussian = true) const
     {
         const float g = genuine(score, gaussian);
@@ -165,7 +186,7 @@ class MatchProbabilityDistance : public Distance
         const QList<int> labels = src.indexProperty(inputVariable);
         QScopedPointer<MatrixOutput> matrixOutput(MatrixOutput::make(FileList(src.size()), FileList(src.size())));
         distance->compare(src, src, matrixOutput.data());
-
+	
         QList<float> genuineScores, impostorScores;
         genuineScores.reserve(labels.size());
         impostorScores.reserve(labels.size()*labels.size());
@@ -178,8 +199,8 @@ class MatchProbabilityDistance : public Distance
                 else                        impostorScores.append(score);
             }
         }
-
-        mp = MP(genuineScores, impostorScores);
+	
+        mp = MP(genuineScores, impostorScores, !gaussian);
     }
 
     float compare(const Template &target, const Template &query) const