Loop closure (LC) plays an important role in Simultaneous Localization and Mapping (SLAM), which can reduce the accumulated drift. This task faces the challenges of large viewpoint changes and expensive computational costs when optimizing the global map. This paper proposes DH-LC, a novel accurate and robust LC method that consists of hierarchical spatial feature matching (HSFM) and hybrid bundle adjustment (HBA). HSFM estimates a reliable relative pose between the query image and the retrieval image in a coarse-to-fine way. Specifically, 3D points are firstly triangulated and then clustered according to the spatial distribution. The cluster centers estimate the coarse matches in a larger perception field which can tolerate large viewpoint changes. HBA optimizes the global map efficiently by adaptively selecting incremental bundle adjustment (IBA) or full bundle adjustment (FBA) according to the accumulated drift and relative pose verification in the temporal window. Experimental results demonstrate that our proposed method easily detects loops in large viewpoint changes and efficiently optimizes the global map. When compared with the state-of-the-art methods, our method increases LC recall and improves SLAM localization accuracy with reducing the accumulated drift.

https://ieeexplore.ieee.org/document/9981061