def detect_emotion_in_frame():
global ct, er
detected_emotions_on_faces = OrderedDict()
request_data = request.get_json()
if request_data["clearGlobals"]:
print("reset signal sent")
ct = CentroidTracker(max_dissapeared=10)
er = EmotionRecognition()
gc.collect()
try:
frame = cv2.imdecode(np.frombuffer(base64.b64decode(
request_data["capturedFrame"][23:]), np.uint8), cv2.IMREAD_COLOR)
except:
return jsonify({"error": "error"})
rects = er.detect_faces(frame)
objects, dissapeared = ct.track(rects)
frame, detected_emotions_on_faces = er.detect_emotions(objects)
flag, encoded_img = cv2.imencode(".jpg", frame)
return jsonify({"frame": base64.b64encode(encoded_img).decode("utf-8"),
"detectedEmotionsOnFaces": list(detected_emotions_on_faces.values()),
"dissapearedFaces": list({"id": d[0], "counter": d[1]} for d in dissapeared.items())})
rects = list()
for i in range(0, detections.shape[2]):
# extract the confidence (i.e., probability) associated with the prediction
confidence = detections[0, 0, i, 2]
# filter out weak detections by ensuring the `confidence` is greater than the minimum confidence
if confidence < 0.6:
continue
# compute the (x, y)-coordinates of the bounding box for the object
box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
rects.append(box.astype("int"))
objects = ct.track(rects)
for (object_id, centroid) in objects.items():
for rect in rects:
(start_x, start_y, end_x, end_y) = rect
c_x = int((start_x + end_x) / 2.0)
c_y = int((start_y + end_y) / 2.0)
if np.array_equal([c_x, c_y], centroid):
face = frame[start_y:end_y, start_x:end_x]
try:
gray_face = cv2.cvtColor(face, cv2.COLOR_BGR2GRAY)
except:
break
