GOSMatch: Graph-of-Semantics Matching for Detecting Loop Closures in 3D LiDAR data

Zhu, Yachen; Ma, Yanyang; Chen, Long; Liu, Cong; Ye, Maosheng; Li, Lingxi

doi:10.1109/iros45743.2020.9341299

Cited by 57 publications

(21 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Readers are referred to [1] for a more thorough review on general loop closure methods. Existing research on 3D-based loop detection can be categorised into three groups [22]: feature-based [7], [8], [11], [23]- [25], segmentation-based [26], [27], and learningbased methods [9], [10], [28], [29].…”

Section: Related Workmentioning

confidence: 99%

“…[32] use semantics from PointNet++ [33] together with the NDT-based histogram descriptors for loop closure detection. GOSMatch [29] further proposse a semantic-level global descriptor that is a histogram-based graph descriptor that encodes relations of objects segmented by RangeNet++ [34]. Instead of performing frame-by-frame loop detection, SegMap [9] segments the scene incrementally as the robot navigates and gives these segments to a deep network as input to generate a signature per segment.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Loop closure detection using local 3D deep descriptors

Zhou¹,

Wang²,

Poiesi³

et al. 2021

Preprint

View full text Add to dashboard Cite

We present a simple yet effective method to address loop closure detection in simultaneous localisation and mapping using local 3D deep descriptors (L3Ds). L3Ds are emerging compact representations of patches extracted from point clouds that are learned from data using a deep learning algorithm. We propose a novel overlap measure for loop detection by computing the metric error between points that correspond to mutually-nearest-neighbour descriptors after registering the loop candidate point cloud by its estimated relative pose. This novel approach enables us to accurately detect loops and estimate six degrees-of-freedom poses in the case of small overlaps. We compare our L3D-based loop closure approach with recent approaches on LiDAR data and achieve state-of-theart loop closure detection accuracy. Additionally, we embed our loop closure approach in RESLAM, a recent edge-based SLAM system, and perform the evaluation on real-world RGBD-TUM and synthetic ICL datasets. Our approach enables RESLAM to achieve a better localisation accuracy compared to its original loop closure strategy.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Loop closure detection using local 3D deep descriptors

Zhou¹,

Wang²,

Poiesi³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In recent years, with the development of point cloud deep learning, many LiDARbased object detection [11] and semantic segmentation [12], [13] methods have been proposed, making it possible to obtain semantic information from point clouds. However, there are still only a few LiDAR-based works trying to use semantic information [7], [14], [15].…”

Section: Introductionmentioning

confidence: 99%

“…Semantic-based methods: SGPR [14] represents the scene as a semantic graph then uses a graph similarity network to score the similarity of the graphs. GOSMatch [15] proposes a new global descriptor that is generated from the spatial relationship between semantics. It also proposes a coarse-to-fine strategy to efficiently search loop closures and gives an accurate 6-DOF initial pose estimation.…”

Section: Introductionmentioning

confidence: 99%

SSC: Semantic Scan Context for Large-Scale Place Recognition

Li¹,

Kong²,

Zhao³

et al. 2021

Preprint

View full text Add to dashboard Cite

Place recognition gives a SLAM system the ability to correct cumulative errors. Unlike images that contain rich texture features, point clouds are almost pure geometric information which makes place recognition based on point clouds challenging. Existing works usually encode low-level features such as coordinate, normal, reflection intensity, etc., as local or global descriptors to represent scenes. Besides, they often ignore the translation between point clouds when matching descriptors. Different from most existing methods, we explore the use of high-level features, namely semantics, to improve the descriptor's representation ability. Also, when matching descriptors, we try to correct the translation between point clouds to improve accuracy. Concretely, we propose a novel global descriptor, Semantic Scan Context, which explores semantic information to represent scenes more effectively. We also present a two-step global semantic ICP to obtain the 3D pose (x, y, yaw) used to align the point cloud to improve matching performance. Our experiments on the KITTI dataset show that our approach outperforms the state-of-theart methods with a large margin. Our code is available at: https://github.com/lilin-hitcrt/SSC.

show abstract

“…the arrangement of certain segments and their spatial relations, hence it is difficult to distinguish scenarios with similar features but different structures. In order to get distinct features, high-level representations such as geometrical segments or semantics were proposed [14], [15], [16], [17], [18]. SegMatch [14] first used Euclidean clustering to segment a point cloud and search candidates by matching K-nearest neighbour (KNN) segments.…”

Section: Introductionmentioning

confidence: 99%

DSC: Deep Scan Context Descriptor for Large-Scale Place Recognition

Cui¹,

Huang²,

Cai³

et al. 2021

Preprint

View full text Add to dashboard Cite

LiDAR-based place recognition is an essential and challenging task both in loop closure detection and global relocalization. We propose Deep Scan Context (DSC), a general and discriminative global descriptor that captures the relationship among segments of a point cloud. Unlike previous methods that utilize either semantics or a sequence of adjacent point clouds for better place recognition, we only use raw point clouds to get competitive results. Concretely, we first segment the point cloud egocentrically to acquire centroids and eigenvalues of the segments. Then, we introduce a graph neural network to aggregate these features into an embedding representation. Extensive experiments conducted on the KITTI dataset show that DSC is robust to scene variants and outperforms existing methods.

show abstract

GOSMatch: Graph-of-Semantics Matching for Detecting Loop Closures in 3D LiDAR data

Cited by 57 publications

References 31 publications

Loop closure detection using local 3D deep descriptors

Loop closure detection using local 3D deep descriptors

SSC: Semantic Scan Context for Large-Scale Place Recognition

DSC: Deep Scan Context Descriptor for Large-Scale Place Recognition

Contact Info

Product

Resources

About