My research interest is at the intersection of machine learning and computer vision, with a particular emphasis on designing efficient inference and learning algorithms for graphical models involved in image and video analysis. My recent projects include visual object detection, tracking and scene segmentation, as well as their applications in bionic eye.