def dnn_detector(frame):
gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
frame_height = frame.shape[0]
frame_width = frame.shape[1]
blob = cv2.dnn.blobFromImage(frame, 1.0, (300, 300), [104, 117, 123],
False, False)
net.setInput(blob)
detections = net.forward()
bboxes = []
idx = 0
offset = 15
x_pos, y_pos = 10, 40
for i in range(detections.shape[2]):
confidence = detections[0, 0, i, 2]
if confidence > conf_threshold:
idx += 1
x1 = int(detections[0, 0, i, 3] * frame_width)
y1 = int(detections[0, 0, i, 4] * frame_height)
x2 = int(detections[0, 0, i, 5] * frame_width)
y2 = int(detections[0, 0, i, 6] * frame_height)
bboxes.append([x1, y1, x2, y2])
face = [x1, y1, x2 - x1, y2 - y1]
if hist_eq:
gray_frame = cv2.equalizeHist(gray_frame)
model_in = get_model_compatible_input(gray_frame,
utils.bb_to_rect(face))
predicted_proba = model.predict(model_in)
predicted_label = np.argmax(predicted_proba[0])
cv2.rectangle(frame, (x1, y1), (x2, y2), (255, 0, 0), 2)
text = f"Person {idx}: {label2text[predicted_label]}"
utils.draw_text_with_backgroud(frame,
text,
x1 + 5,
y1,
font_scale=0.4)
text = f"Person {idx} : "
y_pos = y_pos + 2 * offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
for k, v in label2text.items():
text = f"{v}: {round(predicted_proba[0][k]*100, 3)}%"
y_pos = y_pos + offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
def haar_detector(frame):
gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
face_frame = np.zeros(gray_frame.shape, dtype="uint8")
offset = 15
x_pos, y_pos = 10, 40
faces = cascade_detector.detectMultiScale(gray_frame, 1.32, 5)
for idx, face in enumerate(faces):
if hist_eq:
gray_frame = cv2.equalizeHist(gray_frame)
img_arr = utils.align_face(gray_frame, utils.bb_to_rect(face),
desiredLeftEye)
face_frame = cv2.resize(img_arr, (48, 48),
interpolation=cv2.INTER_CUBIC)
img_arr = utils.preprocess_img(img_arr, resize=False)
predicted_proba = model.predict(img_arr)
predicted_label = np.argmax(predicted_proba[0])
x, y, w, h = face
cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0), 2)
text = f"Person {idx+1}: {label2text[predicted_label]}"
utils.draw_text_with_backgroud(frame, text, x + 5, y, font_scale=0.4)
text = f"Person {idx+1} : "
y_pos = y_pos + 2 * offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
for k, v in label2text.items():
text = f"{v}: {round(predicted_proba[0][k]*100, 3)}%"
y_pos = y_pos + offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
return frame, face_frame
def dlib_detector(frame_orig):
frame = frame_orig.copy()
gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
offset = 15
x_pos, y_pos = 10, 40
faces = hog_detector(gray_frame)
for idx, face in enumerate(faces):
if hist_eq:
gray_frame = cv2.equalizeHist(gray_frame)
img_arr = utils.align_face(gray_frame, face, desiredLeftEye)
img_arr = utils.preprocess_img(img_arr, resize=False)
predicted_proba = model.predict(img_arr)
predicted_label = np.argmax(predicted_proba[0])
x, y, w, h = rect_to_bb(face)
cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0), 2)
text = f"Person {idx+1}: {label2text[predicted_label]}"
utils.draw_text_with_backgroud(frame, text, x + 5, y, font_scale=0.4)
text = f"Person {idx+1} : "
y_pos = y_pos + 2 * offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
for k, v in label2text.items():
text = f"{v}: {round(predicted_proba[0][k]*100, 3)}%"
y_pos = y_pos + offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
return frame
def dlib_detector(frame):
gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
offset = 15
x_pos, y_pos = 10, 40
faces = hog_detector(gray_frame)
for idx, face in enumerate(faces):
if hist_eq:
gray_frame = cv2.equalizeHist(gray_frame)
model_in = get_model_compatible_input(gray_frame, face)
predicted_proba = model.predict(model_in)
predicted_label = np.argmax(predicted_proba[0])
x, y, w, h = rect_to_bb(face)
cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0), 2)
text = f"Person {idx+1}: {label2text[predicted_label]}"
utils.draw_text_with_backgroud(frame, text, x + 5, y, font_scale=0.4)
text = f"Person {idx+1} : "
y_pos = y_pos + 2 * offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
for k, v in label2text.items():
text = f"{v}: {round(predicted_proba[0][k]*100, 3)}%"
y_pos = y_pos + offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
if iswebcam:
frame = cv2.flip(frame, 1, 0)
try:
tik = time.time()
if args["detector"] == "dlib":
out = dlib_detector(frame)
else:
out = dnn_detector(frame)
tt += time.time() - tik
fps = frame_count / tt
label = f"Detector: {args['detector']} ; Model: {args['model']}; HistEq: {args['histogram_equalization']} ; FPS: {round(fps, 2)}"
utils.draw_text_with_backgroud(frame,
label,
10,
20,
font_scale=0.35)
except Exception as e:
print(e)
pass
cv2.imshow("Face Detection Comparison", frame)
if cv2.waitKey(10) == ord('q'):
break
cv2.destroyAllWindows()
vidcap.release()
def dnn_detector(frame):
frame_height = frame.shape[0]
frame_width = frame.shape[1]
blob = cv2.dnn.blobFromImage(frame, 1.0, (300, 300), [104, 117, 123],
False, False)
net.setInput(blob)
detections = net.forward()
bboxes = []
idx = 0
offset = 15
x_pos, y_pos = 10, 40
gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
face_frame = np.zeros(gray_frame.shape, dtype="uint8")
for i in range(detections.shape[2]):
confidence = detections[0, 0, i, 2]
if confidence > conf_threshold:
idx += 1
x1 = int(detections[0, 0, i, 3] * frame_width)
y1 = int(detections[0, 0, i, 4] * frame_height)
x2 = int(detections[0, 0, i, 5] * frame_width)
y2 = int(detections[0, 0, i, 6] * frame_height)
bboxes.append([x1, y1, x2, y2])
face = [x1, y1, x2 - x1, y2 - y1]
if hist_eq:
gray_frame = cv2.equalizeHist(gray_frame)
img_arr = utils.align_face(gray_frame, utils.bb_to_rect(face),
desiredLeftEye)
face_frame = cv2.resize(img_arr, (48, 48),
interpolation=cv2.INTER_CUBIC)
img_arr = utils.preprocess_img(img_arr, resize=False)
predicted_proba = model.predict(img_arr)
predicted_label = np.argmax(predicted_proba[0])
cv2.rectangle(frame, (x1, y1), (x2, y2), (255, 0, 0), 2)
text = f"Person {idx}: {label2text[predicted_label]}"
utils.draw_text_with_backgroud(frame,
text,
x1 + 5,
y1,
font_scale=0.4)
text = f"Person {idx} : "
y_pos = y_pos + 2 * offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
for k, v in label2text.items():
text = f"{v}: {round(predicted_proba[0][k]*100, 3)}%"
y_pos = y_pos + offset
utils.draw_text_with_backgroud(frame,
text,
x_pos,
y_pos,
font_scale=0.3,
box_coords_2=(2, -2))
return frame, face_frame
img = cv2.resize(thresh, (120, 120), fx=0.5, fy=0.5)
img = np.expand_dims(img, axis=2)
img = np.expand_dims(img, axis=0)
img = img / 255.
predicted_proba = model.predict(img)
predicted_label = np.argmax(predicted_proba[0])
bb_text = label2text[predicted_label]
except Exception as e:
bb_text = "no hand"
if high_resolution:
utils.draw_text_with_backgroud(frame,
bb_text,
x=start_coords[0],
y=start_coords[1],
font_scale=1.2)
tt += time.time() - tik
fps = round(frame_count / tt, 2)
main_text = "Running..." + f" fps: {fps}"
utils.draw_text_with_backgroud(frame,
main_text,
x=15,
y=25,
font_scale=1.,
thickness=2)
utils.draw_text_with_backgroud(frame,
"Instructions for better results:",
x=15,
if args["mode"] == "debug":
combined = np.hstack((bb, thresh))
cv2.imshow("bb v/s thresh", combined)
img = cv2.resize(thresh, (120, 120))
img = np.expand_dims(img, axis=2)
img = np.expand_dims(img, axis=0)
img = img / 255.
predicted_proba = model.predict(img)
predicted_label = np.argmax(predicted_proba[0])
bb_text = label2text[predicted_label]
except Exception as e:
bb_text = "no hand"
utils.draw_text_with_backgroud(frame, bb_text, x=start_coords[0], y=start_coords[1], font_scale=0.4)
tt += time.time() - tik
fps = round(frame_count / tt, 2)
main_text = "Go On..." + f" fps: {fps}"
utils.draw_text_with_backgroud(frame, main_text, x=15, y=25, font_scale=0.35, thickness=1)
utils.draw_text_with_backgroud(frame, "Instructions for better results :-", x=15, y=55, font_scale=0.32, thickness=1)
utils.draw_text_with_backgroud(frame, "- Place your hand completely inside the window", x=15, y=75, font_scale=0.32, thickness=1)
utils.draw_text_with_backgroud(frame, "- Place your hand close to window", x=15, y=95, font_scale=0.32, thickness=1)
else:
main_text = "Within 5 seconds ensure that the background behind the window doesn't change"
utils.draw_text_with_backgroud(frame, main_text, x=15, y=25, font_scale=0.35, thickness=1)
cv2.imshow("out", frame)
if cv2.waitKey(10) == ord("q"):
break
break
frame_count += 1
if iswebcam:
frame = cv2.flip(frame, 1, 0)
try:
tik = time.time()
out_dlib = dlib_detector(frame)
tt_dlib += time.time() - tik
fps_dlib = frame_count / tt_dlib
label = f"Detector: dlib ; HistEq: {args['histogram_equalization']} ; FPS: {round(fps_dlib, 2)}"
utils.draw_text_with_backgroud(out_dlib,
label,
10,
20,
font_scale=0.35)
tik = time.time()
out_dnn = dnn_detector(frame)
tt_dnn += time.time() - tik
fps_dnn = frame_count / tt_dnn
label = f"Detector: dnn_tf ; HistEq: {args['histogram_equalization']} ; FPS: {round(fps_dnn, 2)}"
utils.draw_text_with_backgroud(out_dnn,
label,
10,
20,
font_scale=0.35)
frame = np.hstack([out_dlib, out_dnn])