Merge pull request #16 from thoughtworksarts/pr-13/sinbycos/ferWebcamOpenCV

Impoves on PR #13 : Webcam example

Author: xuv

.gitignore

@@ -13,3 +13,4 @@ trained_models/
 venv/
 output/
 emopy_venv/
+EmoPy/examples/image_data/image.jpg

EmoPy.egg-info/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.1
 Name: EmoPy
-Version: 0.0.2
+Version: 0.0.4
 Summary: A deep neural net toolkit for emotion analysis via Facial Expression Recognition (FER)
 Home-page: https://github.com/thoughtworksarts/EmoPy
 Author: ThoughtWorks Arts
-Author-email: andy@thoughtworks.io
+Author-email: info@thoughtworksarts.io
 License: UNKNOWN
 Description: # EmoPy
         EmoPy is a python toolkit with deep neural net classes which accurately predict emotions given images of people's faces.
@@ -85,13 +85,12 @@ Description: # EmoPy
 
         ## Installation
 
-        <!--
+        
         ### From PyPi
         Once the virtual environment is activated, you may install EmoPy using
         ```
         pip install EmoPy
         ```
-        -->
 
         ### From the source
 
@@ -113,26 +112,38 @@ Description: # EmoPy
 
         ## Running the examples
 
-        You can find example code to run each of the current neural net classes in [examples](examples). The best place to start is the [FERModel example](examples/fermodel_example.py). Here is a listing of that code:
+        You can find example code to run each of the current neural net classes in [examples](examples). You may either download the example directory to a location of your choice on your machine, or find the example directory included in the installation.
+        
+        If you choose to use the installed package, you can find the examples directory by starting in the virtual environment directory you created and typing:
+        ```
+        cd lib/python3.6/site-packages/EmoPy/examples
+        ```
+        
+        
+        The best place to start is the [FERModel example](examples/fermodel_example.py). Here is a listing of that code:
 
         ```python
-        import sys
-        sys.path.append('../')
-        from fermodel import FERModel
+        from EmoPy.src.fermodel import FERModel
+        from pkg_resources import resource_filename
 
         target_emotions = ['calm', 'anger', 'happiness']
         model = FERModel(target_emotions, verbose=True)
 
         print('Predicting on happy image...')
-        model.predict('image_data/sample_happy_image.png')
+        model.predict(resource_filename('EmoPy.examples','image_data/sample_happy_image.png'))
+        
+        print('Predicting on disgust image...')
+        model.predict(resource_filename('EmoPy.examples','image_data/sample_disgust_image.png'))
+        
+        print('Predicting on anger image...')
+        model.predict(resource_filename('EmoPy.examples','image_data/sample_anger_image2.png'))
         ```
 
         The code above loads a pre-trained model and then predicts an emotion on a sample image. As you can see, all you have to supply with this example is a set of target emotions and a sample image.
 
-        Once you have completed the installation, you can run this example by moving into the examples folder and running the example script.
+        Once you have completed the installation, you can run this example from the examples folder by running the example script.
 
         ```
-        cd examples
         python fermodel_example.py
         ```
 
@@ -143,7 +154,6 @@ Description: # EmoPy
         To train your own neural net, use one of our FER neural net classes to get started. You can try the convolutional_model.py example:
 
         ```
-        cd examples
         python convolutional_example.py
         ```
 
@@ -224,5 +234,7 @@ Description: # EmoPy
         [@vanGent2016]: http://www.paulvangent.com/2016/04/01/emotion-recognition-with-python-opencv-and-a-face-dataset/ "Emotion Recognition With Python, OpenCV and a Face Dataset. A tech blog about fun things with Python and embedded electronics."
 
 Platform: UNKNOWN
-Requires-Python: >=3.6.3,<3.7
+Classifier: Programming Language :: Python :: 3.6
+Classifier: Operating System :: MacOS :: MacOS X
+Requires-Python: >=3.6.3
 Description-Content-Type: text/markdown

EmoPy.egg-info/SOURCES.txt

@@ -15,6 +15,7 @@ EmoPy/examples/__init__.py
 EmoPy/examples/convolutional_lstm_model.py
 EmoPy/examples/convolutional_model.py
 EmoPy/examples/fermodel_example.py
+EmoPy/examples/fermodel_example_webcam.py
 EmoPy/examples/timedelay_conv_model.py
 EmoPy/examples/transferlearning_model.py
 EmoPy/examples/image_data/sample.csv

EmoPy/examples/fermodel_example_webcam.py

@@ -2,40 +2,43 @@
 
 import cv2
 import sys
-sys.path.append('../')
+from EmoPy.src.fermodel import FERModel
+from pkg_resources import resource_filename
 
-#Choose the type of Face Expression Model
-from src.fermodel import FERModel
+fontFace = cv2.FONT_HERSHEY_SIMPLEX;
+fontScale = 1;
+thickness = 2;
 
-#Frame Number
-FRAME_NUM = 0
-
-#Choose the type of face detector cascade you want to use
-cascPath = "~/EmoPy/venv/lib/python3.5/site-packages/cv2/data/haarcascade_frontalface_default.xml"
-faceCascade = cv2.CascadeClassifier(cascPath)
 #Specify the camera which you want to use. The default argument is '0'
 video_capture = cv2.VideoCapture(0)
+#Capturing a smaller image fçor speed purposes
+video_capture.set(cv2.CAP_PROP_FRAME_WIDTH, 640)
+video_capture.set(cv2.CAP_PROP_FRAME_HEIGHT, 360)
+video_capture.set(cv2.CAP_PROP_FPS, 15)
+
+#Can choose other target emotions from the emotion subset defined in fermodel.py in src directory. The function
+# defined as `def _check_emotion_set_is_supported(self):`
+target_emotions = ['calm', 'anger', 'happiness']
+model = FERModel(target_emotions, verbose=True)
 
 while True:
 	#Capture frame-by-frame
 	ret, frame = video_capture.read()
 	#Save the captured frame on disk
-	file = '~/EmoPy/models/examples/image_data/image.jpg'
+	file = 'image_data/image.jpg'
 	cv2.imwrite(file, frame)
-	#Can choose other target emotions from the emotion subset defined in fermodel.py in src directory. The function
-	# defined as `def _check_emotion_set_is_supported(self):`
-	target_emotions = ['calm', 'anger', 'happiness']
-	model = FERModel(target_emotions, verbose=True)
+
 	frameString = model.predict(file)
-	#Display frame number and emotion 
-	cv2.putText(frame, 'Frame:' + str(FRAME_NUM), (10, 40), cv2.FONT_HERSHEY_COMPLEX_SMALL, 3, (0, 0, 255), 2, cv2.LINE_AA)
-	cv2.putText(frame, frameString, (10,450), cv2.FONT_HERSHEY_COMPLEX_SMALL, 3, (0,255,0), 2, cv2.LINE_AA)
+
+	#Display emotion
+	retval, baseline = cv2.getTextSize(frameString, fontFace, fontScale, thickness)
+	cv2.rectangle(frame, (0, 0 ), (20 + retval[0], 50 ), (0,0,0), -1 )
+	cv2.putText(frame, frameString, (10, 35), fontFace, fontScale, (255, 255, 255), thickness, cv2.LINE_AA)
 	cv2.imshow('Video', frame)
 	cv2.waitKey(1)
-	FRAME_NUM += 1
+
 	#Press Esc to exit the window
 	if cv2.waitKey(1) & 0xFF == 27:
 		break
 #Closes all windows
 cv2.destroyAllWindows()
-

EmoPy/src/fermodel.py

@@ -3,6 +3,7 @@
 from scipy import misc
 import numpy as np
 import json
+from pkg_resources import resource_filename
 
 class FERModel:
     """
@@ -56,7 +57,7 @@ def predict(self, image_file):
         resized_image = cv2.resize(gray_image, self.target_dimensions, interpolation=cv2.INTER_LINEAR)
         final_image = np.array([np.array([resized_image]).reshape(list(self.target_dimensions)+[self.channels])])
         prediction = self.model.predict(final_image)
-        ### Return the dominant expression
+        # Return the dominant expression
         dominant_expression = self._print_prediction(prediction[0])
         return dominant_expression
 
@@ -93,10 +94,10 @@ def _choose_model_from_target_emotions(self):
         sorted_indices = [str(idx) for idx in sorted(model_indices)]
         model_suffix = ''.join(sorted_indices)
         #Modify the path to choose the model file and the emotion map that you want to use
-        model_file = '~/EmoPy/models/conv_model_%s.hdf5' % model_suffix
-        emotion_map_file = '~/EmoPy/models/conv_emotion_map_%s.json' % model_suffix
-        emotion_map = json.loads(open(emotion_map_file).read())
-        return load_model(model_file), emotion_map
+        model_file = 'models/conv_model_%s.hdf5' % model_suffix
+        emotion_map_file = 'models/conv_emotion_map_%s.json' % model_suffix
+        emotion_map = json.loads(open(resource_filename('EmoPy', emotion_map_file)).read())
+        return load_model(resource_filename('EmoPy', model_file)), emotion_map
 
     def _print_prediction(self, prediction):
         normalized_prediction = [x/sum(prediction) for x in prediction]
@@ -110,4 +111,3 @@ def _print_prediction(self, prediction):
         # print('Dominant emotion: %s' % dominant_emotion)
         # print()
         return dominant_emotion
-

README.md

@@ -38,14 +38,10 @@ Predictions ideally perform well on a diversity of datasets, illumination condit
 
 ## Environment Setup
 
-EmoPy runs using Python 3.6, theoretically on any Python-compatible OS. We tested EmoPy using Python 3.6.6 on OSX. You can install [Python 3.6.6](https://www.python.org/downloads/release/python-366/) from the Python website.
-
-Please note that this is not the most current version of Python, but the TensorFlow package doesn't work with Python 3.7 yet, so EmoPy cannot run with Python 3.7.
+EmoPy runs using Python 3.6 and up, theoretically on any Python-compatible OS. We tested EmoPy using Python 3.6.6 on OSX. You can install [Python 3.6.6](https://www.python.org/downloads/release/python-366/) from the Python website.
 
 Python is compatible with multiple operating systems. If you would like to use EmoPy on another OS, please convert these instructions to match your target environment. Let us know how you get on, and we will try to support you and share you results.
 
-
-
 If you do not have Homebrew installed run this command to install:
 
 ```
@@ -67,7 +63,7 @@ Create and activate the virtual environment. Run:
 ```
 python3.6 -m venv venv
 ```
-where the second `venv` is the name of your virtual environment. To activate, run from the same directory: 
+where the second `venv` is the name of your virtual environment. To activate, run from the same directory:
 ```
 source venv/bin/activate
 ```

setup.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="EmoPy",
-    version="0.0.4",
+    version="0.0.5",
     author="ThoughtWorks Arts",
     author_email="[email protected]",
     description="A deep neural net toolkit for emotion analysis via Facial Expression Recognition (FER)",
@@ -18,7 +18,7 @@
         "Programming Language :: Python :: 3.6",
         "Operating System :: MacOS :: MacOS X"
     ],
-    python_requires='>=3.6.3,<3.7',
+    python_requires='>=3.6.3',
     install_requires=[
        'keras>=2.2.0',
        'lasagne',