Filteryaseen

filters/mouse.py

@@ -0,0 +1,73 @@
+import cv2
+import numpy as np
+import dlib
+from math import hypot
+filter_image = cv2.imread("assest\ghoul2.png")
+filter_image3 = cv2.imread("assest/hair9.png")
+filter_image4 = cv2.imread("assest/eye1.png")
+filter_image5 = cv2.imread("assest/eye2.png")
+
+
+# Loading Face detector
+detector = dlib.get_frontal_face_detector()
+predictor = dlib.shape_predictor("assest\shape_predictor_68_face_landmarks.dat")
+
+def filteringmouse(cap,rows, cols):
+    filter1 = np.zeros((rows, cols), np.uint8)
+    _, frame = cap.read()
+    filter1.fill(0)
+    gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+    faces = detector(frame)
+
+    # if faces:
+    try:
+        filter(frame,gray_frame,faces,filter_image,66,66,1,1,20)
+        # filter(frame, gray_frame, faces, filter_image3,27,27,1.5,2,25)
+
+        filter(frame, gray_frame, faces, filter_image3,27,27,1.5,1,100)
+        filter(frame, gray_frame, faces, filter_image4,40,40,0.2,0.4,5)
+        filter(frame, gray_frame, faces, filter_image5,46,46,0.2,0.4,5)
+
+
+
+    except:
+        _, frame_f = cap.read()
+        cv2.imshow("Frame", frame_f)
+    # else:
+    #     _, frame_f = cap.read()
+    #     cv2.imshow("Frame", frame_f)
+def filter(frame,gray_frame,faces,filter_image1,X,Y,width,height,above=0,left=0):
+    for face in faces:
+        landmarks = predictor(gray_frame, face)
+
+        # filter coordinates
+        # top_filter = (landmarks.part(27).x+10, landmarks.part(24).y+10)
+        center_filter = (landmarks.part(X).x-left, landmarks.part(Y).y-above)
+        left_filter = (landmarks.part(4).x, landmarks.part(4).y)
+        right_filter = (landmarks.part(14).x, landmarks.part(14).y)
+
+        filter_width = int(hypot(left_filter[0] - right_filter[0],
+                                 left_filter[1] - right_filter[1]) * width)
+        filter_height = int(filter_width * height)
+
+        # New filter position
+        top_left = (int(center_filter[0] - filter_width / 2),
+                    int(center_filter[1] - filter_height / 2))
+        bottom_right = (int(center_filter[0] + filter_width / 2),
+                        int(center_filter[1] + filter_height / 2))
+
+        # Adding the new filter
+        # coloring
+        filtery = cv2.resize(filter_image1, (filter_width, filter_height))
+        filtery_gray = cv2.cvtColor(filtery, cv2.COLOR_BGR2GRAY)
+        _, filter1 = cv2.threshold(filtery_gray, 25, 255, cv2.THRESH_BINARY_INV)
+
+        filter_area = frame[top_left[1]: top_left[1] + filter_height,
+                      top_left[0]: top_left[0] + filter_width]
+        filter_area_no_filter = cv2.bitwise_and(filter_area, filter_area, mask=filter1)
+        final_filter = cv2.add(filter_area_no_filter, filtery)
+
+        frame[top_left[1]: top_left[1] + filter_height,
+        top_left[0]: top_left[0] + filter_width,:] = final_filter
+
+        cv2.imshow("Frame", frame)

main.py

@@ -2,15 +2,19 @@
 import numpy as np
 import dlib
 from filters.filter1 import filtering
-from filters.dogfilter import filteringdog
+# from filters.dogfilter import filteringdog
+from filters.mouse import filteringmouse
+
 # Loading Camera and Nose image and Creating mask
 cap = cv2.VideoCapture(0)
+cap.set(3, 1920)
+cap.set(4, 1080)
 _, frame = cap.read()
 # print(frame)
 rows, cols, _ = frame.shape
 filter1 = np.zeros((rows, cols), np.uint8)
 while True:
-    filteringdog(cap,rows, cols)
+    filteringmouse(cap,rows, cols)
     key = cv2.waitKey(1)
     if key == 27:
         break