def initialize(self, base_info, diff_data, game_setting):
self.base_info = base_info
self.game_setting = game_setting
# seed設定
random.seed(random.random() + self.base_info["agentIdx"])
self.myrole = base_info["myRole"]
self.result_seer = []
self.result_med = []
self.player_size = len(self.base_info["remainTalkMap"].keys())
self.talk_turn = 0
self.honest = False
self.divined_as_wolf = []
self.divined_as_human = []
self.wrong_divine = set()
self.white = set()
self.black = set()
self.greys = set()
self.check_alive()
self.seers = set()
self.tryingPP = set()
self.greys = set(self.alive) - {int(base_info["agentIdx"])}
self.players = self.greys.copy()
self.whisper_turn = 0
self.attack_success = True
self.attacked_who_lastnight = 0
self.gen.gameInitialize(self.base_info["agentIdx"], self.player_size)
#self.q = [] # 廃止しました
self.seer_divined_me_as_werewolf = set()
self.estimated_me_as_werewolf = set()
self.estimated_me_as_human = set()
self.asked_why_divine = set()
self.asked_why_doubt = set()
self.PPmode = False
self.when_declare = random.randint(4, 8)
self.emo = emotion.Emotion(self.alive)
self.stealth = False if random.random() < 0.75 else True
self.who_said_black = dict(zip(self.alive, [[] for i in self.alive]))
self.has_CO_seer = False
self.emo.myrole_appearance = base_info["myRole"]
self.emo.myrole = base_info["myRole"]
def __init__(self, canvas, media_module, media_file):
self.vid = emotion.Emotion(canvas, module_filename=media_module)
self.member_add(self.vid)
self.vid.file = media_file
self.vid.smooth_scale = True
self.frame = edje.Edje(canvas,
file=theme_file,
size=(320, 240),
group="video_controller")
self.member_add(self.frame)
self.frame.part_swallow("video_swallow", self.vid)
self.frame.data["moving"] = False
self.frame.data["resizing"] = False
self.frame.part_drag_value_set("video_speed", 0.0, 1.0)
self.frame.part_text_set("video_speed_txt", "1.0")
self.vid.show()
evas.SmartObject.__init__(self, canvas)
self._setup_signals()
def main():
# Detect Face factor
minsize = 20 # minimum size of face
threshold = [0.6, 0.7, 0.8] # three steps's threshold
factor = 0.709 # scale factor
print('Creating networks and loading parameters')
# with tf.Graph().as_default():
# sess = tf.Session()
# with sess.as_default():
# pnet, rnet, onet = align.detect_face.create_mtcnn(sess, None)
with tf.Graph().as_default():
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
with sess.as_default():
pnet, rnet, onet = align.detect_face.create_mtcnn(sess, None)
# Load emotion
emotion = em.Emotion()
emotion.load_weights("models/emotion/emotion.h5")
# Camera
capture = cv2.VideoCapture(0)
if not capture:
print("Can't get image from webcam")
sys.exit(0)
frame_interval = 3
count = 0
while True:
cam_status, frame = capture.read()
if not cam_status:
print(cam_status)
break
if count == 0:
img = frame
bounding_boxes, _ = align.detect_face.detect_face(
img, minsize, pnet, rnet, onet, threshold, factor)
# if bounding_boxes.shape[0] <= 0:
# continue
# Prediction and Draw rectangle
for face_position in bounding_boxes:
face_position = face_position.astype(int)
# Predict emotion
face = crop_and_resize(img,
(face_position[0], face_position[1],
face_position[2], face_position[3]))
emotion_result = emotion.predict(face)
emotion_max = max(emotion_result, key=emotion_result.get)
face_emotion = "{} {:.5f}".format(emotion_max,
emotion_result[emotion_max])
# Draw rectangle at the face position
cv2.rectangle(frame, (face_position[0], face_position[1]),
(face_position[2], face_position[3]),
(0, 255, 0), 2)
cv2.putText(frame, face_emotion,
(face_position[0], face_position[1] - 20),
cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (0, 255, 0))
cv2.imshow('Webcam', frame)
else:
pass
count += 1
if count >= frame_interval:
count = 0
if cv2.waitKey(1) & 0xFF == ord('q'):
break
capture.release()
cv2.destroyAllWindows()
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=0.75, allow_growth=True)
tf_config = tf.ConfigProto(
gpu_options=gpu_options, log_device_placement=False)
sess = tf.Session(config=tf_config)
with sess.as_default():
pnet, rnet, onet = align.detect_face.create_mtcnn(sess, None)
facenet.load_model(os.path.abspath("./models/facenet/20180402-114759/"))
images_placeholder = tf.get_default_graph().get_tensor_by_name("input:0")
embeddings = tf.get_default_graph().get_tensor_by_name("embeddings:0")
phase_train_placeholder = tf.get_default_graph().get_tensor_by_name("phase_train:0")
# Load emotion
emotion = em.Emotion()
emotion.load_weights("models/emotion/emotion.h5")
# Load Facenet
classifier_filename_exp = os.path.abspath("./models/facenet/20180402-114759/lfw_classifier.pkl")
with open(classifier_filename_exp, 'rb') as infile:
(model, class_names) = pickle.load(infile)
def load_facenet_model():
classifier_filename_exp = os.path.abspath("./models/facenet/20180402-114759/lfw_classifier.pkl")
with open(classifier_filename_exp, 'rb') as infile:
(model, class_names) = pickle.load(infile)
def crop_and_resize(image, position, size):
""" Crop Function take path, x1, y1, x2, y2 then give back cropped photo """
posx1, posy1, posx2, posy2 = position
def cozmo_manage_emotion(robot: cozmo.robot.Robot):
cozmo_robot = emotion.Emotion(robot)
robot_plot = pd.DataFrame(cozmo_robot.emotions, index=[0])
print("Welcome to the emotional manager!")
print("Please select a emotion.")
results_view_loop(cozmo_robot, robot_plot)