diff --git a/examples/assignment_learning_ex.cpp b/examples/assignment_learning_ex.cpp
index a7103c67a..bafc5a098 100644
--- a/examples/assignment_learning_ex.cpp
+++ b/examples/assignment_learning_ex.cpp
@@ -195,13 +195,10 @@ int main()
 
 
         // Finally, the assigner can be serialized to disk just like most dlib objects.
-        ofstream fout("assigner.dat", ios::binary);
-        serialize(assigner, fout);
-        fout.close();
+        serialize("assigner.dat") << assigner;
 
         // recall from disk
-        ifstream fin("assigner.dat", ios::binary);
-        deserialize(assigner, fin);
+        deserialize("assigner.dat") >> assigner;
     }
     catch (std::exception& e)
     {
diff --git a/examples/custom_trainer_ex.cpp b/examples/custom_trainer_ex.cpp
index ec3fc550b..39af53f39 100644
--- a/examples/custom_trainer_ex.cpp
+++ b/examples/custom_trainer_ex.cpp
@@ -208,15 +208,12 @@ int main()
             decision_function<radial_basis_kernel<sample_type> >  // This is the output of the rbf_trainer
         > df2, df3;
 
-
     df2 = df;
-    ofstream fout("df.dat", ios::binary);
-    serialize(df2, fout);
-    fout.close();
+    // save to a file called df.dat
+    serialize("df.dat") << df2;
 
     // load the function back in from disk and store it in df3.  
-    ifstream fin("df.dat", ios::binary);
-    deserialize(df3, fin);
+    deserialize("df.dat") >> df3;
 
 
     // Test df3 to see that this worked.
diff --git a/examples/fhog_object_detector_ex.cpp b/examples/fhog_object_detector_ex.cpp
index d2d246b96..6ddb02aa3 100644
--- a/examples/fhog_object_detector_ex.cpp
+++ b/examples/fhog_object_detector_ex.cpp
@@ -180,14 +180,11 @@ int main(int argc, char** argv)
 
         // Like everything in dlib, you can save your detector to disk using the
         // serialize() function.
-        ofstream fout("face_detector.svm", ios::binary);
-        serialize(detector, fout);
-        fout.close();
+        serialize("face_detector.svm") << detector;
 
         // Then you can recall it using the deserialize() function.
-        ifstream fin("face_detector.svm", ios::binary);
         object_detector<image_scanner_type> detector2;
-        deserialize(detector2, fin);
+        deserialize("face_detector.svm") >> detector2;
 
 
 
diff --git a/examples/krls_ex.cpp b/examples/krls_ex.cpp
index d20b90777..968f1a6dd 100644
--- a/examples/krls_ex.cpp
+++ b/examples/krls_ex.cpp
@@ -78,21 +78,17 @@ int main()
 
     // Another thing that is worth knowing is that just about everything in dlib is serializable.
     // So for example, you can save the test object to disk and recall it later like so:
-    ofstream fout("saved_krls_object.dat",ios::binary);
-    serialize(test,fout);
-    fout.close();
+    serialize("saved_krls_object.dat") << test;
 
     // Now let's open that file back up and load the krls object it contains.
-    ifstream fin("saved_krls_object.dat",ios::binary);
-    deserialize(test, fin);
+    deserialize("saved_krls_object.dat") >> test;
 
     // If you don't want to save the whole krls object (it might be a bit large) 
     // you can save just the decision function it has learned so far.  You can get 
     // the decision function out of it by calling test.get_decision_function() and
     // then you can serialize that object instead.  E.g.
     decision_function<kernel_type> funct = test.get_decision_function();
-    fout.open("saved_krls_function.dat",ios::binary);
-    serialize(funct, fout);
+    serialize("saved_krls_function.dat") << funct;
 }
 
 
diff --git a/examples/krr_classification_ex.cpp b/examples/krr_classification_ex.cpp
index 4eff004c0..42648351f 100644
--- a/examples/krr_classification_ex.cpp
+++ b/examples/krr_classification_ex.cpp
@@ -196,14 +196,10 @@ int main()
 
     // Another thing that is worth knowing is that just about everything in dlib is serializable.
     // So for example, you can save the learned_pfunct object to disk and recall it later like so:
-    ofstream fout("saved_function.dat",ios::binary);
-    serialize(learned_pfunct,fout);
-    fout.close();
+    serialize("saved_function.dat") << learned_pfunct;
 
     // Now let's open that file back up and load the function object it contains.
-    ifstream fin("saved_function.dat",ios::binary);
-    deserialize(learned_pfunct, fin);
-
+    deserialize("saved_function.dat") >> learned_pfunct;
 
 }
 
diff --git a/examples/krr_regression_ex.cpp b/examples/krr_regression_ex.cpp
index f1bb23694..26c1412d7 100644
--- a/examples/krr_regression_ex.cpp
+++ b/examples/krr_regression_ex.cpp
@@ -94,14 +94,10 @@ int main()
 
     // Another thing that is worth knowing is that just about everything in dlib is serializable.
     // So for example, you can save the test object to disk and recall it later like so:
-    ofstream fout("saved_function.dat",ios::binary);
-    serialize(test,fout);
-    fout.close();
+    serialize("saved_function.dat") << test;
 
     // Now let's open that file back up and load the function object it contains.
-    ifstream fin("saved_function.dat",ios::binary);
-    deserialize(test, fin);
-
+    deserialize("saved_function.dat") >> test;
 
 }
 
diff --git a/examples/learning_to_track_ex.cpp b/examples/learning_to_track_ex.cpp
index 55e79cc16..2f9f3947f 100644
--- a/examples/learning_to_track_ex.cpp
+++ b/examples/learning_to_track_ex.cpp
@@ -344,13 +344,10 @@ int main()
 
 
     // Finally, you can save your track_association_function to disk like so:
-    ofstream fout("track_assoc.svm", ios::binary);
-    serialize(assoc, fout);
-    fout.close();
+    serialize("track_assoc.svm") << assoc;
 
     // And recall it from disk later like so:
-    ifstream fin("track_assoc.svm", ios::binary);
-    deserialize(assoc, fin);
+    deserialize("track_assoc.svm") >> assoc;
 }
 
 // ----------------------------------------------------------------------------------------
diff --git a/examples/multiclass_classification_ex.cpp b/examples/multiclass_classification_ex.cpp
index cb7c46abd..782511ca3 100644
--- a/examples/multiclass_classification_ex.cpp
+++ b/examples/multiclass_classification_ex.cpp
@@ -138,13 +138,10 @@ int main()
         // Put df into df2 and then save df2 to disk.  Note that we could have also said
         // df2 = trainer.train(samples, labels);  But doing it this way avoids retraining.
         df2 = df;
-        ofstream fout("df.dat", ios::binary);
-        serialize(df2, fout);
-        fout.close();
+        serialize("df.dat") << df2;
 
         // load the function back in from disk and store it in df3.  
-        ifstream fin("df.dat", ios::binary);
-        deserialize(df3, fin);
+        deserialize("df.dat") >> df3;
 
 
         // Test df3 to see that this worked.
diff --git a/examples/object_detector_ex.cpp b/examples/object_detector_ex.cpp
index f7168fe62..cda71eb5a 100644
--- a/examples/object_detector_ex.cpp
+++ b/examples/object_detector_ex.cpp
@@ -247,13 +247,10 @@ int main()
 
 
         // Finally, note that the detector can be serialized to disk just like other dlib objects.
-        ofstream fout("object_detector.dat", ios::binary);
-        serialize(detector, fout);
-        fout.close();
+        serialize("object_detector.dat") << detector;
 
         // Recall from disk.
-        ifstream fin("object_detector.dat", ios::binary);
-        deserialize(detector, fin);
+        deserialize("object_detector.dat") >> detector;
     }
     catch (exception& e)
     {
diff --git a/examples/rvm_ex.cpp b/examples/rvm_ex.cpp
index 7ed04b724..d1d5935e7 100644
--- a/examples/rvm_ex.cpp
+++ b/examples/rvm_ex.cpp
@@ -208,14 +208,10 @@ int main()
 
     // Another thing that is worth knowing is that just about everything in dlib is serializable.
     // So for example, you can save the learned_pfunct object to disk and recall it later like so:
-    ofstream fout("saved_function.dat",ios::binary);
-    serialize(learned_pfunct,fout);
-    fout.close();
+    serialize("saved_function.dat") << learned_pfunct;
 
     // Now let's open that file back up and load the function object it contains.
-    ifstream fin("saved_function.dat",ios::binary);
-    deserialize(learned_pfunct, fin);
-
+    deserialize("saved_function.dat") >> learned_pfunct;
 
 }
 
diff --git a/examples/rvm_regression_ex.cpp b/examples/rvm_regression_ex.cpp
index c2604b583..d65cb5203 100644
--- a/examples/rvm_regression_ex.cpp
+++ b/examples/rvm_regression_ex.cpp
@@ -91,14 +91,10 @@ int main()
 
     // Another thing that is worth knowing is that just about everything in dlib is serializable.
     // So for example, you can save the test object to disk and recall it later like so:
-    ofstream fout("saved_function.dat",ios::binary);
-    serialize(test,fout);
-    fout.close();
+    serialize("saved_function.dat") << test;
 
     // Now let's open that file back up and load the function object it contains.
-    ifstream fin("saved_function.dat",ios::binary);
-    deserialize(test, fin);
-
+    deserialize("saved_function.dat") >> test;
 
 }
 
diff --git a/examples/sequence_labeler_ex.cpp b/examples/sequence_labeler_ex.cpp
index eee78ab57..bdb666a7e 100644
--- a/examples/sequence_labeler_ex.cpp
+++ b/examples/sequence_labeler_ex.cpp
@@ -292,13 +292,10 @@ int main()
 
 
     // Finally, the labeler can be serialized to disk just like most dlib objects.
-    ofstream fout("labeler.dat", ios::binary);
-    serialize(labeler, fout);
-    fout.close();
+    serialize("labeler.dat") << labeler;
 
     // recall from disk
-    ifstream fin("labeler.dat", ios::binary);
-    deserialize(labeler, fin);
+    deserialize("labeler.dat") >> labeler;
 }
 
 // ----------------------------------------------------------------------------------------
diff --git a/examples/sequence_segmenter_ex.cpp b/examples/sequence_segmenter_ex.cpp
index ba0897d48..3b0eb8cde 100644
--- a/examples/sequence_segmenter_ex.cpp
+++ b/examples/sequence_segmenter_ex.cpp
@@ -228,13 +228,10 @@ int main()
 
 
     // Finally, the segmenter can be serialized to disk just like most dlib objects.
-    ofstream fout("segmenter.dat", ios::binary);
-    serialize(segmenter, fout);
-    fout.close();
+    serialize("segmenter.dat") << segmenter;
 
     // recall from disk
-    ifstream fin("segmenter.dat", ios::binary);
-    deserialize(segmenter, fin);
+    deserialize("segmenter.dat") >> segmenter;
 }
 
 // ----------------------------------------------------------------------------------------
diff --git a/examples/svm_ex.cpp b/examples/svm_ex.cpp
index 7d598536f..3d5d0bb84 100644
--- a/examples/svm_ex.cpp
+++ b/examples/svm_ex.cpp
@@ -210,13 +210,10 @@ int main()
     // Another thing that is worth knowing is that just about everything in dlib is
     // serializable.  So for example, you can save the learned_pfunct object to disk and
     // recall it later like so:
-    ofstream fout("saved_function.dat",ios::binary);
-    serialize(learned_pfunct,fout);
-    fout.close();
+    serialize("saved_function.dat") << learned_pfunct;
 
     // Now let's open that file back up and load the function object it contains.
-    ifstream fin("saved_function.dat",ios::binary);
-    deserialize(learned_pfunct, fin);
+    deserialize("saved_function.dat") >> learned_pfunct;
 
     // Note that there is also an example program that comes with dlib called the
     // file_to_code_ex.cpp example.  It is a simple program that takes a file and outputs a
diff --git a/examples/train_object_detector.cpp b/examples/train_object_detector.cpp
index b861ad20f..6ba478490 100644
--- a/examples/train_object_detector.cpp
+++ b/examples/train_object_detector.cpp
@@ -291,9 +291,7 @@ int main(int argc, char** argv)
                 object_detector<image_scanner_type> detector = trainer.train(images, object_locations, ignore);
 
                 cout << "Saving trained detector to object_detector.svm" << endl;
-                ofstream fout("object_detector.svm", ios::binary);
-                serialize(detector, fout);
-                fout.close();
+                serialize("object_detector.svm") << detector;
 
                 cout << "Testing detector on training data..." << endl;
                 cout << "Test detector (precision,recall,AP): " << test_object_detection_function(detector, images, object_locations) << endl;