Methods for Appearance-based Loop Closure Detection

Addresses the vision-based loop closure detection problem from different points of view
Deals with the problem of generating topological maps of the environment using efficient appearance-based loop closure detection techniques
Proposes a Bag of Binary Words (BoBW) approach, which generates the dictionary online, avoiding the classical training step, and this scheme is used to efficiently detect loop closures in the image stream

Inhalt
Introduction.- Background.- Literature Review.- Experimental Setup.- Loop Closure Detection using Local Features and KD-Trees.- Loop Closure Detection using Incremental Bags of Binary Words.- Hierarchical Loop Closure Detection for Topological Mapping.- Fast Image Mosaicking using Incremental Bags of Binary Words.- Conclusions and Future Work.