HBST: A Hamming Distance embedding Binary Search Tree for Visual Place Recognition
This work addresses the need for reliable and efficient visual place recognition in SLAM systems, presenting an incremental improvement over prior methods.
The authors tackled the problem of efficient binary descriptor matching and image retrieval for visual place recognition by introducing HBST, a Hamming distance embedding binary search tree, which achieves descriptor search and insertion in logarithmic time and demonstrates competitive performance on public datasets.
Reliable and efficient Visual Place Recognition is a major building block of modern SLAM systems. Leveraging on our prior work, in this paper we present a Hamming Distance embedding Binary Search Tree (HBST) approach for binary Descriptor Matching and Image Retrieval. HBST allows for descriptor Search and Insertion in logarithmic time by exploiting particular properties of binary Feature descriptors. We support the idea behind our search structure with a thorough analysis on the exploited descriptor properties and their effects on completeness and complexity of search and insertion. To validate our claims we conducted comparative experiments for HBST and several state-of-the-art methods on a broad range of publicly available datasets. HBST is available as a compact open-source C++ header-only library.