def get_mood(self, uid): picture_model_service = PictureModelService() picture = picture_model_service.find_by_key(uid) if picture is not None and picture.mood_process is not None: return picture.mood_process else: return []
def perfom_mood_detection(self, uid): picture_model_service = PictureModelService() picture = picture_model_service.find_by_key(uid) if picture is None or picture.image_path is None: return [] mood_process = MoodProcess() mood_process.startDate = datetime.datetime.utcnow() # extract features and landmarks feature_landmarks = self.extractor.extract_with_landmark(picture.image_path) # for each feature/landmark predict mood and convert landmark to points for feature_landmark in feature_landmarks: face = Face() mood = self.classifier.predict(feature_landmark.features) face.mood = const.moods[int(mood[0])] face.points = self.convert2points(feature_landmark.landmark) mood_process.faces.append(face) mood_process.endDate = datetime.datetime.utcnow() picture.mood_process = mood_process # save picture model picture_model_service.save(picture) return picture.mood_process
def perfom_mood_detection(self, uid): picture_model_service = PictureModelService() picture = picture_model_service.find_by_key(uid) if picture is None or picture.image_path is None: return [] mood_process = MoodProcess() mood_process.startDate = datetime.datetime.utcnow() # extract features and landmarks feature_landmarks = self.extractor.extract_with_landmark( picture.image_path) # for each feature/landmark predict mood and convert landmark to points for feature_landmark in feature_landmarks: face = Face() mood = self.classifier.predict(feature_landmark.features) face.mood = const.moods[int(mood[0])] face.points = self.convert2points(feature_landmark.landmark) mood_process.faces.append(face) mood_process.endDate = datetime.datetime.utcnow() picture.mood_process = mood_process # save picture model picture_model_service.save(picture) return picture.mood_process