Bag-of-Word-Groups (BoWG): A Robust and Efficient Loop Closure Detection Method Under Perceptual Aliasing

TL;DR

BoWG tackles loop closure under perceptual aliasing in SLAM by introducing word groups that capture co-occurrence and proximity of visual words and by embedding temporal consistency directly into similarity scoring. It builds an online BoW-like dictionary with word-group and inverse/direct indices, plus a feature distribution module and post-verification. Empirical results on public datasets and a pipe-domain dataset show BoWG achieves higher precision-recall and CPU efficiency (average $16\,\mathrm{ms}$ per image on CPU for $17{,}565$ images) than state-of-the-art methods, demonstrating strong scalability in large-scale applications. The approach is particularly effective in narrow, texture-repetitive environments, indicating significant practical impact for robust SLAM in challenging settings.

Abstract

Loop closure is critical in Simultaneous Localization and Mapping (SLAM) systems to reduce accumulative drift and ensure global mapping consistency. However, conventional methods struggle in perceptually aliased environments, such as narrow pipes, due to vector quantization, feature sparsity, and repetitive textures, while existing solutions often incur high computational costs. This paper presents Bag-of-Word-Groups (BoWG), a novel loop closure detection method that achieves superior precision-recall, robustness, and computational efficiency. The core innovation lies in the introduction of word groups, which captures the spatial co-occurrence and proximity of visual words to construct an online dictionary. Additionally, drawing inspiration from probabilistic transition models, we incorporate temporal consistency directly into similarity computation with an adaptive scheme, substantially improving precision-recall performance. The method is further strengthened by a feature distribution analysis module and dedicated post-verification mechanisms. To evaluate the effectiveness of our method, we conduct experiments on both public datasets and a confined-pipe dataset we constructed. Results demonstrate that BoWG surpasses state-of-the-art methods, including both traditional and learning-based approaches, in terms of precision-recall and computational efficiency. Our approach also exhibits excellent scalability, achieving an average processing time of 16 ms per image across 17,565 images in the Bicocca25b dataset.

Figures (10)

• Figure 1: The proposed Bag-of-Word-Groups (BoWG) method enhances loop closure detection by integrating multiple information modalities for similarity computation. The F1 score reflects the precision and recall performance. Learning-based methods (highlighted in blue) were implemented on GPU, while our systems achieves outstanding F1 score with superior execution speed on CPU. In addition, Our system significantly improves the discrimination of the true loop closure area under perceptual aliasing.
• Figure 2: Workflow of the proposed BoWG loop closure detection method. BoWG method consists of five main modules: Feature Extraction, Word Processing, Word Group Processing, Distribution Processing, and Verification.
• Figure 3: Some examples of word groups.
• Figure 4: The designed database of our BoWG system.
• Figure 5: The proposed feature distribution analysis method.