add demo

Author: jrobchin

.gitignore

@@ -3,4 +3,5 @@
 training_data/*
 validation_data/*
 
-*.whl
+*.whl
+pyenv/

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+    "python.pythonPath": "pyenv/bin/python3"
+}

__pycache__/utils.cpython-36.pyc

@@ -0,0 +1,101 @@
+import sys # system functions (ie. exiting the program)
+import os # operating system functions (ie. path building on Windows vs. MacOs)
+import time # for time operations
+import uuid # for generating unique file names
+import math # math functions
+import argparse
+
+from IPython.display import display as ipydisplay, Image, clear_output, HTML # for interacting with the notebook better
+
+import numpy as np # matrix operations (ie. difference between two matricies)
+import cv2 # (OpenCV) computer vision functions (ie. tracking)
+(major_ver, minor_ver, subminor_ver) = (cv2.__version__).split('.')
+print('OpenCV Version: {}.{}.{}'.format(major_ver, minor_ver, subminor_ver))
+
+from utils import setup_tracker
+
+# Helper function for applying a mask to an array
+def mask_array(array, imask):
+    if array.shape[:2] != imask.shape:
+        raise Exception("Shapes of input and imask are incompatible")
+    output = np.zeros_like(array, dtype=np.uint8)
+    for i, row in enumerate(imask):
+        output[i, row] = array[i, row]
+    return output
+
+
+# Begin capturing video
+video = cv2.VideoCapture(0)
+if not video.isOpened():
+    print("Could not open video")
+    sys.exit()
+
+
+# Read first frame
+ok, frame = video.read()
+if not ok:
+    print("Cannot read video")
+    sys.exit()
+# Use the first frame as an initial background frame
+bg = frame.copy()
+
+
+# Kernel for erosion and dilation of masks
+kernel = np.ones((3,3),np.uint8)
+
+
+# Capture, process, display loop    
+while True:
+    # Read a new frame
+    ok, frame = video.read()
+    display = frame.copy()
+    if not ok:
+        break
+        
+        
+    # Start timer
+    timer = cv2.getTickCount()
+
+    # Processing
+    # First find the absolute difference between the two images
+    diff = cv2.absdiff(bg, frame)
+    mask = cv2.cvtColor(diff, cv2.COLOR_BGR2GRAY)
+    # Threshold the mask
+    th, thresh = cv2.threshold(mask, 10, 255, cv2.THRESH_BINARY)
+    # Opening, closing and dilation
+    opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
+    closing = cv2.morphologyEx(opening, cv2.MORPH_CLOSE, kernel)
+    img_dilation = cv2.dilate(closing, kernel, iterations=2)
+    # Get mask indexes
+    imask = img_dilation > 0
+    # Get foreground from mask
+    foreground = mask_array(frame, imask)
+    
+        
+    # Calculate Frames per second (FPS)
+    fps = cv2.getTickFrequency() / (cv2.getTickCount() - timer)
+    # Display FPS on frame
+    
+    
+    # Display diff
+    cv2.imshow("diff", diff)
+    # Display thresh
+    cv2.imshow("thresh", thresh)
+    # Display mask
+    cv2.imshow("img_dilation", img_dilation)
+    # Display foreground
+    cv2.imshow("foreground", foreground)
+    
+    cv2.putText(display, "FPS : " + str(int(fps)), (100, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.65, (50, 170, 50), 2)
+    # Display result
+    cv2.imshow("original", display)
+    
+    k = cv2.waitKey(1) & 0xff
+    if k == 27: break # ESC pressed
+    elif k == 114 or k == 112: 
+        # r pressed
+        bg = frame.copy()
+    elif k != 255: print(k)
+        
+cv2.destroyAllWindows()
+video.release()

demos/__pycache__/utils.cpython-36.pyc

@@ -18,6 +18,12 @@ class GestureRecognizer:
         3: 'swing'
     }
 
+    positions = {
+        'hand_pose': (15, 40), # hand pose text
+        'fps': (15, 20), # fps counter
+        'null_pos': (200, 200) # used as null point for mouse control
+    }
+
     def __init__(self, capture=0, model_path=MODEL_PATH):
         self.bg = None
         self.frame = None # current frame
@@ -27,21 +33,14 @@ def __init__(self, capture=0, model_path=MODEL_PATH):
         self.is_tracking = False
         self.hand_bbox = (116, 116, 170, 170)
 
-        # Display positions (pixel coordinates)
-        self.positions = {
-            'hand_pose': (15, 40), # hand pose text
-            'fps': (15, 20), # fps counter
-            'null_pos': (200, 200) # used as null point for mouse control
-        }
-
         # Begin capturing video
-        self.video = cv2.VideoCapture(1)
+        self.video = cv2.VideoCapture(capture)
         if not self.video.isOpened():
             print("Could not open video")
 
-    def __del__(self):
-        cv2.destroyAllWindows()
-        self.video.release()
+    # def __del__(self):
+    #     cv2.destroyAllWindows()
+    #     self.video.release()
 
     # Helper function for applying a mask to an array
     def mask_array(self, array, imask):
@@ -71,7 +70,7 @@ def extract_foreground(self):
         # Get foreground from mask
         foreground = self.mask_array(self.frame, imask)
 
-        return foreground, mask
+        return foreground, img_dilation
 
     def run(self):
         # Capture, process, display loop    
@@ -210,6 +209,7 @@ def run(self):
                 # Display foreground_display
                 cv2.imshow("foreground_display", foreground_display)
                 cv2.imshow("display", display)
+                cv2.imshow("hand_crop", hand_crop)
                 # Display result
                 cv2.imshow("data", data_display)
 
@@ -221,6 +221,7 @@ def run(self):
                     self.is_tracking = False
                 elif k == 116: # t pressed
                     # Initialize tracker with first frame and bounding box
+                    self.is_tracking = True
                     self.tracker = cv2.TrackerKCF_create()
                     self.tracker.init(self.frame, self.hand_bbox)
                 elif k == 112: # p pressed
@@ -231,5 +232,5 @@ def run(self):
                 elif k != 255: print(k)
 
 if __name__ == "__main__":
-    recognizer = GestureRecognizer(1)
+    recognizer = GestureRecognizer(0)
     recognizer.run()