Kyung-shik Roh

Particle filter-based heading estimation using magnetic compasses for mobile robot navigation

Heading information is critical for the control and/or navigation of mobile devices and robots. To get accurate heading information robustly, we propose a method which combines particle filtering with magnetic compasses. Although magnetic compasses can provide absolute heading angle, they have not been used for indoor applications since serious magnetic interferences are commonly founded in home/office environments. We overcome this difficulty by 1) suggesting statistical calibration of a magnetic compass, 2) deriving necessary conditions of the Earths magnetic field area, and 3) designing an event-based particle filter based on likelihood calculated from conditional probability. Particle filter is an emerging key technology which can be applied to nonlinear/non-Gaussian model, beyond the limitations of Kalman filter. We take advantage of particle filter to optimally fuse the information from both magnetic compasses and odometry data. Experimental results on mobile robot navigation in indoor environments show reliability and robustness of the proposed method

Markov Network-Based Unified Classifier for Face Identification

We propose a novel unifying framework using a Markov network to learn the relationship between multiple classifiers in face recognition. We assume that we have several complementary classifiers and assign observation nodes to the features of a query image and hidden nodes to the features of gallery images. We connect each hidden node to its corresponding observation node and to the hidden nodes of other neighboring classifiers. For each observation-hidden node pair, we collect a set of gallery candidates that are most similar to the observation instance, and the relationship between the hidden nodes is captured in terms of the similarity matrix between the collected gallery images. Posterior probabilities in the hidden nodes are computed by the belief-propagation algorithm. The novelty of the proposed framework is the method that takes into account the classifier dependency using the results of each neighboring classifier. We present extensive results on two different evaluation protocols, known and unknown image variation tests, using three different databases, which shows that the proposed framework always leads to good accuracy in face recognition.