Robust Loop Closure Detection based on Bag of SuperPoints and Graph Verification

Yue, Haosong; Miao, Jinyu; Yu, Yue; Chen, Weihai; Chen, Wen

doi:10.1109/iros40897.2019.8967726

Cited by 30 publications

(27 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the incorporation of lines in an LCD method not only improves the performance in low-textured scenarios, but for these low-textured datasets, and provides a competitive performance for the remaining datasets. Furthermore, LiPo-LCD ++ obtains better recall values almost for every dataset compared to the solution by Yue et al (2019), which describes the images using a non-binarized version of the SuperPoint descriptor. Despite LiPo-LCD ++ achieves better recall values than our previous solution in the MLG dataset, the performance attained is rather low with regard to other more successful approaches.…”

Section: Comparison With Other Solutionsmentioning

confidence: 97%

“…Due to these reasons, point local descriptors, either real-valued (Cummins and Newman 2008;Angeli et al 2008;Cummins and Newman 2011;Tsintotas et al 2019) or binary (Galvez-López and Tardos 2012; Mur-Artal and Tardós 2014; Khan and Wollherr 2015;Garcia-Fidalgo and Ortiz 2018), have been widely used in the literature during last decades. More recently, approaches based on CNNs have emerged as an alternative, motivated by their demonstrated robustness to visual appearance changes (Chen et al 2014;Sünderhauf et al 2015;Chen et al 2017;Kenshimov et al 2017;Lopez-Antequera et al 2017;Yue et al 2019). These solutions typically employ CNNs to extract a global descriptor of the image, what makes more difficult the implementation of a spatial verification stage.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Appearance-based loop closure detection combining lines and learned points for low-textured environments

2022

View full text Add to dashboard Cite

Hand-crafted point descriptors have been traditionally used for visual loop closure detection. However, in low-textured environments, it is usually difficult to find enough point features and, hence, the performance of such algorithms degrade. Under this context, this paper proposes a loop closure detection method that combines lines and learned points to work, particularly, in scenarios where hand-crafted points fail. To index previous images, we adopt separate incremental binary Bag-of-Words (BoW) schemes for points and lines. Moreover, we adopt a binarization procedure for features’ descriptors to benefit from the advantages of learned features into a binary BoW model. Furthermore, image candidates from each BoW instance are merged using a novel query-adaptive late fusion approach. Finally, a spatial verification stage, which integrates appearance and geometry perspectives, allows us to enhance the global performance of the method. Our approach is validated using several public datasets, outperforming other state-of-the-art solutions in most cases, especially in low-textured scenarios.

show abstract

Section: Comparison With Other Solutionsmentioning

confidence: 97%

Section: Related Workmentioning

confidence: 99%

Appearance-based loop closure detection combining lines and learned points for low-textured environments

2022

View full text Add to dashboard Cite

show abstract

“…There are two main challenges to perform visual place recognition based on the differences of scenes: one is the appearance change caused by illumination condition and seasonal changes, and the other is the viewpoint change caused by revisiting one place from different viewpoints [1]. In the VPR literature, various feature extraction methods have been developed for visual place recognition, including deep convolutional feature-based methods [6][7][8][9][10][11][12][13], handicraft feature-based methods [2,18], semantic information-based methods [19][20][21][22][23][24][25], sequence-based methods [26,27], and graph-based methods [19,20,[28][29][30][31][32]. Overall, most of these studies focus on the image processing module of the visual place recognition system, which aims to extract and describe features that are robust in the different challenge conditions as mentioned above.…”

Section: Visual Place Recognitionmentioning

confidence: 99%

GSAP: A Global Structure Attention Pooling Method for Graph-Based Visual Place Recognition

Yang

Liu

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

The Visual Place Recognition problem aims to use an image to recognize the location that has been visited before. In most of the scenes revisited, the appearance and view are drastically different. Most previous works focus on the 2-D image-based deep learning method. However, the convolutional features are not robust enough to the challenging scenes mentioned above. In this paper, in order to take advantage of the information that helps the Visual Place Recognition task in these challenging scenes, we propose a new graph construction approach to extract the useful information from an RGB image and a depth image and fuse them in graph data. Then, we deal with the Visual Place Recognition problem as a graph classification problem. We propose a new Global Pooling method—Global Structure Attention Pooling (GSAP), which improves the classification accuracy by improving the expression ability of the Global Pooling component. The experiments show that our GSAP method improves the accuracy of graph classification by approximately 2–5%, the graph construction method improves the accuracy of graph classification by approximately 4–6%, and that the whole Visual Place Recognition model is robust to appearance change and view change.

show abstract

“…In this case, we need to consider the repeatability and discriminability of features. Secondly, the traditional BoW framework displaces the spatial information between visual words, resulting in quantization errors [ 12 ]. Sparse feature matching used to be solved with hand-crafted descriptors [ 13 , 14 , 15 ].…”

Section: Introductionmentioning

confidence: 99%

“…Recently, Convolutional Neural Networks (CNN) has had great success in pattern recognition and computer vision tasks [ 16 , 17 ]. Many researchers use the depth features extracted by CNN to improve the LCD algorithm and achieve fine results [ 12 , 18 , 19 ]. However, the calculation amount of LCD based on CNN is quite enormous.…”

Section: Introductionmentioning

confidence: 99%

LFM: A Lightweight LCD Algorithm Based on Feature Matching between Similar Key Frames

Zhu

Liu

et al. 2021

Sensors

View full text Add to dashboard Cite

Loop Closure Detection (LCD) is an important technique to improve the accuracy of Simultaneous Localization and Mapping (SLAM). In this paper, we propose an LCD algorithm based on binary classification for feature matching between similar images with deep learning, which greatly improves the accuracy of LCD algorithm. Meanwhile, a novel lightweight convolutional neural network (CNN) is proposed and applied to the target detection task of key frames. On this basis, the key frames are binary classified according to their labels. Finally, similar frames are input into the improved lightweight feature matching network based on Transformer to judge whether the current position is loop closure. The experimental results show that, compared with the traditional method, LFM-LCD has higher accuracy and recall rate in the LCD task of indoor SLAM while ensuring the number of parameters and calculation amount. The research in this paper provides a new direction for LCD of robotic SLAM, which will be further improved with the development of deep learning.

show abstract

Robust Loop Closure Detection based on Bag of SuperPoints and Graph Verification

Cited by 30 publications

References 35 publications

Appearance-based loop closure detection combining lines and learned points for low-textured environments

Appearance-based loop closure detection combining lines and learned points for low-textured environments

GSAP: A Global Structure Attention Pooling Method for Graph-Based Visual Place Recognition

LFM: A Lightweight LCD Algorithm Based on Feature Matching between Similar Key Frames

Contact Info

Product

Resources

About