“…The application of visual compass can be classified into different types based on various criteria, including type of used sensor, method of estimation, and type of used features. Depending on the used sensors, visual compasses can be classified as monocular cameras [1][2][3], stereo cameras [4][5][6], RGB-D camera [7][8][9], polarized light camera [10][11][12], omnidirectional cameras [3,13,14], or a combination of sensors [15,16]. Meanwhile, visual compasses have different estimation methods by focusing on various types of information, such as 2D-3D correspondence [17][18][19], virtual visual servoing (VVS) [20][21][22], holistic method [23][24][25], extended Kalman filter (EKF) [26][27][28], and appearance-based method [3,13,29].…”