Learning, positioning, and tracking visual appearance | IEEE Conference Publication | IEEE Xplore