FAB-MAP: Appearance-Based Place Recognition and Mapping using a Learned Visual Vocabulary Model.

Mark Joseph Cummins, Paul M. Newman

Browse the full ICML paper archive.