Emotion¶

Emotion description

Functions

void *rf_emotion_init(const char *model_path, rf_error_type *error)¶

Initializes emotion handle.

Return

emotion handle

Parameters

model_path: path to the “emotion” model
error: output parameter to return an error details

void rf_emotion_close(void *classifier_handle, rf_error_type *error)¶

Deletes emotion handle.

Parameters

classifier_handle: classifier handle for the emotion
error: output parameter to return an error details

int rf_emotion_classify_whole(void *classifier_handle, void *image_handle, int ymin, int xmin, int ymax, int xmax, rf_error_type *error)¶

Classify whole image’s emotion (where the whole image is the face image after cropped)

Parameters

classifier_handle: classifier handle
image_handle: image handle (use image API to create)
error: output parameter to return an error details output: int which represents the face’s emotion using the following mapping: denote the output emotion as: output_emotion and the mapping by emotions_table_by_indices: std::vector<std::string> emotions_table_by_indices = {std::string(“angry”), std::string(“disgust”), std::string(“fear”), std::string(“happy”), std::string(“sad”), std::string(“surprise”), std::string(“neutral”) }; So the output emotion as string is given by: emotions_table_by_indices.at(output_emotion)