LiDAR Loop Closure Detection using Semantic Graphs with Graph Attention Networks

Ignacio Alzugaray; Liudi Yang; Marco Karrer; Margarita Chli; Ruben Mascaro; Sai Manoj Prakhya; Ziyuan Liu

REVIEW 4 major objections 6 minor 42 references

LiDAR Loop Closure Detection using Semantic Graphs with Graph Attention Networks

T0 review · 4 major / 6 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims that a semantic graph encoder built on graph attention networks, combined with a comparison module based on the difference of graph vectors, detects LiDAR loop closures more accurately than previous semantic-graph place…

desk verdict Solid incremental upgrade to SGPR with real, mostly believable gains; the 13%/19% headline needs explicit baseline provenance and error bars before it is fully trustable. read the letter →

arxiv 2501.19382 v1 pith:5F2JMZXS submitted 2025-01-31 cs.CV cs.RO

Liudi Yang , Ruben Mascaro , Ignacio Alzugaray , Sai Manoj Prakhya , Marco Karrer , Ziyuan Liu , Margarita Chli This is my paper · ORCID

classification cs.CVcs.RO

keywords LiDARloopclosuresemanticgraphsgraphattentionnetworksplacerecognitionregistrationself-attentionSLAMpointcloud

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper is trying to establish that a robot can recognize a previously visited place from a sparse semantic graph of the objects around it, using graph attention networks to turn that graph into a short descriptor vector. This matters for LiDAR SLAM, where recognizing a revisiting place lets the system correct accumulated drift. The proposed pipeline encodes each object's semantic label, centroid, and bounding box into a 32-dimensional graph vector, then classifies a pair of scans as the same place or not by comparing their vectors with a learned difference-based similarity module. On SemanticKITTI the method reaches a maximum F1 of 0.921, and on KITTI-360 0.842, which the paper reports as a 13% and 19% improvement over the SGPR baseline. A semantic registration step then estimates the 6-DoF relative pose from candidate loop scans and inserts it as a constraint into a pose-graph SLAM system.

What carries the argument

The central objects are the semantic graph, whose nodes carry a one-hot semantic label, a centroid coordinate, and a six-dimensional bounding box, and the graph-attention encoder that processes it. The encoder runs three graph-attention branches over a k-nearest-neighbor neighborhood with k equal to 10, concatenates the branch outputs, fuses them with a self-attention layer into a node embedding, and then uses a second self-attention layer to learn a global context vector that weights nodes into a 32-dimensional graph vector. The comparison module computes a similarity vector from the absolute difference of two graph vectors using first-order and second-order difference terms together with the concatenated vectors, and a binary classifier decides whether the scan pair is a loop closure. This machinery is what converts raw semantic instances into a compact, comparable place signature.

What would settle it

Repeat the SemanticKITTI evaluation on scans with more than 50 instances, re-running the random node sampling several times per scan pair; if the spread in max F1 across re-samplings is comparable to the reported 13% improvement over SGPR, then the random sampling assumption is doing much of the work and the gain may not transfer to denser scenes.

Watch

Extended reading notes

Core claim

The central claim is that lightweight attention over semantic graphs is a stronger representation of a place than the edge-convolution graph embedding used by SGPR. By replacing EdgeConv with multi-head graph attention, adding bounding-box geometry to node features, fusing node branches with self-attention, and introducing the absolute difference of two graph vectors into the similarity network, the paper raises mean max F1 on SemanticKITTI from 0.814 (SGPR) to 0.921 and on KITTI-360 from 0.705 to 0.842. The difference term alone contributes the largest single gain, moving the baseline from 0.814 to 0.876 in the ablation study. The paper also shows the approach is robust to random yaw rotation and occlusion, degrades only mildly when semantic labels come from a pretrained segmentation network rather than ground truth, and runs at about 73 Hz with a 426 KB model. The same semantic information feeds a registration step that estimates the 6-DoF pose constraint for the SLAM system.

Load-bearing premise

The method assumes that a sparse graph of up to 50 object instances with semantic labels, centroids, and bounding boxes is a stable and discriminative signature of a place, so that two scans of the same place yield similar graphs even when segmentation is imperfect and nodes are randomly sampled.

Editorial extensions

If this is right

If the central claim holds, semantic-graph loop closure no longer needs dense point-cloud descriptors: a 426 KB model running at about 73 Hz is enough to detect revisiting places on standard benchmarks.
The reported 13% and 19% max-F1 gains over SGPR imply that graph attention plus explicit difference-based comparison is a direct upgrade path for existing semantic-graph place recognition systems.
The robustness results imply that loop closure can survive rotation and occlusion better than Scan Context and Intensity Scan Context, since the semantic graph abstracts away raw point density and local surface details.
The semantic registration results imply that label-aware point-to-line and point-to-plane alignment with dynamic-object removal can serve as both a front-end odometry and a loop-closure pose estimator, with ATE improved on most tested SemanticKITTI sequences.
The strong results on sequence 08, where the robot revisits places in the reverse direction, suggest that semantic topology captured by graph attention helps disambiguate direction, a case where classical descriptors often fail.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

One extension the paper leaves implicit is that the difference-vector comparison term could improve other graph-similarity tasks, such as object-graph retrieval or scene-graph matching in images, wherever pairs of graph embeddings are compared.
The paper only evaluates a fixed set of semantic categories; a testable extension is whether the same graph encoder works with class-agnostic instance proposals or noisy open-vocabulary labels, since the graph representation is designed to tolerate node-level label noise.
Because bounding boxes outperformed FPFH and PointNet features despite carrying less geometric detail, a testable inference is that object scale and rough boundaries are the discriminative cues for place identity; this could be checked by ablating each bounding-box dimension.
The random sampling of nodes above 50 is a stability concern: measuring the variance of max F1 across repeated samplings on scans with many instances would show how much of the reported gain depends on the particular 50-node subset.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: this is a competent, incremental extension of SGPR for semantic-graph loop closure, and the empirical gains look real. The authors swap SGPR's EdgeConv for GAT encoding, add bounding-box geometry to the graph nodes, and add a difference-vector term to the graph comparison module. None of those pieces is new by itself, but the combination works and the paper demonstrates it with a proper ablation and open-source code.

Credit where due: the ablation in Table 5 is the backbone of the paper. Each component adds a stepwise F1 gain on SemanticKITTI (0.814 to 0.876 to 0.882 to 0.908 to 0.921), so the improvement is not a black-box jump. The model is tiny (426 KB) and runs at 73 Hz, which matters for SLAM integration. The robustness experiments under rotation and occlusion, and the full SLAM integration where ISC-LOAM fails on sequence 02, are useful additions. The authors also acknowledge the KITTI-360 label-recovery noise up front.

The soft spots are real but fixable. First, the paper never states whether the SGPR baseline numbers in Table 1 come from re-running the official SGPR code under the same pair-generation, node cap, semantic remapping, and train/test splits, or from the original SGPR paper. That matters for the headline 13%/19% claim. I would not call this fatal: the mean gap is large (0.107 on SemanticKITTI) and the ablation independently corroborates the architecture's value, so I doubt the gain is an artifact. But the authors need to state the baseline provenance and ideally publish the re-run. Second, all results are single runs with max F1 selected on the test set, and the pipeline has stochastic pieces (random node sampling above 50 instances, random pair sampling, random initialization). The paper should report variance across seeds. Given the size of the gaps I expect the conclusion survives, but I cannot verify that from the paper.

Minor: the semantic registration weights (1.2 for signs/poles/buildings, 0.8 otherwise) look hand-set, and the claim that graph-vector dimension 16–64 makes no noticeable difference is stated without numbers.

Who it is for: people working on LiDAR place recognition or semantic-graph SLAM. It is a solid comparison point and a reasonable baseline for future work. It deserves a serious referee; I would ask for baseline re-run details and variance statistics in the revision, then accept.

Referee Report

4 major / 6 minor

Summary. The paper proposes a LiDAR loop-closure detection pipeline based on semantic graphs. A semantic graph encoder uses three graph-attention branches over semantic labels, centroids, and bounding boxes, fuses the branches with self-attention, and pools the node embedding into a 32-dimensional graph vector. A graph comparison module classifies scan pairs using a similarity vector that includes the absolute difference between the two graph vectors, and a semantic registration module estimates the 6-DoF pose between candidate scans. The authors evaluate on SemanticKITTI, KITTI-360, and RangeNet++-labeled KITTI, reporting max F1 of 0.921 and 0.842 versus SGPR's 0.814 and 0.705, which they phrase as 13% and 19% improvements, together with ablations, robustness experiments, and SLAM integration.

Significance. If the results hold, the paper is a useful engineering contribution: it replaces SGPR's EdgeConv encoder with graph attention networks, adds bounding-box geometry, and shows that a difference-based comparison module is effective. The model is very lightweight (426 KB, 73 Hz on a T4), the evaluation covers multiple datasets and ablations, and the code is open-source. The design is internally coherent, and I found no circularity: the network is trained and tested on external public datasets, and the global context vector in Eq. (5) is data-dependent rather than a hand-fitted parameter. However, the headline gains over SGPR rest on baseline provenance and single-run max-F1 comparisons that are not yet documented precisely enough, so the numerical claims are currently under-specified.

major comments (4)

[Section 4.1.3, Table 1] The manuscript does not state whether the SGPR and SGPR-RN rows were produced by re-running the official SGPR implementation under exactly the same evaluation protocol as Ours (same data splits, 3 m/20 m pair-generation thresholds, 50-node cap, semantic class remapping, training/test sequences, and candidate generation), or whether the numbers were taken from the original publication. Since the 13% and 19% improvement over SGPR is the paper's central claim, please document the baseline setup, re-run the baselines under identical conditions, and report any values taken from the literature separately with appropriate caveats.
[Section 4.1.2, Table 1] All results in Table 1 are single-run max-F1 values selected on the test set, even though the pipeline contains stochastic components: random node sampling when a scan has more than 50 instances, random pair sampling, and random weight initialization. The KITTI-360 mean gap over SGPR is 0.137, and the KITTI-360 labels were recovered by clustering. Without multiple seeds and a dispersion measure (mean ± std across seeds, or bootstrap confidence intervals), the reported differences cannot be separated from run-to-run variation; the threshold-selection procedure should also be specified (validation set versus test set).
[Section 4.1.3, KITTI-360 paragraph] The paper acknowledges that KITTI-360 labels were recovered from submap annotations by an unspecified clustering method, introducing 'some annotation errors/noise.' Because KITTI-360 is one of the two datasets supporting the headline 19% improvement, please describe the clustering procedure, quantify its accuracy on a validation subset, and report a sensitivity analysis, for example by comparing results on high-confidence labels only or on scans not affected by clustering artifacts.
[Section 4.1.4 versus Table 1] Section 4.1.4 states that the Ours-RN and SGPR-RN rows in Table 1 refer to performance on the KITTI dataset with RangeNet++ labels, but Table 1 as printed contains only SemanticKITTI and KITTI-360 columns, with no KITTI-RN column. The claimed 'about 5% max F1 score' drop under real-world label inference is therefore not verifiable from the reported table; please correct the table or the text and report the KITTI-RN results explicitly.

minor comments (6)

[Section 3.2.1] Please specify how object instances are obtained from per-point semantic labels and how the bounding boxes are computed; 'readily available through instance segmentation' is not sufficient to reproduce the semantic graph construction.
[Section 3.2.2, Eq. (3)] The pooling operation used to obtain the node embedding f in Eq. (3) is not defined; please state whether it is mean pooling, max pooling, or another reduction.
[Section 3.2.2, Eq. (1)] The k-nearest-neighbor neighborhood Ni is not fully specified: it should be stated in which feature space the search is performed, whether it is computed separately for each branch, and how zero-feature pseudo nodes are treated.
[Table 5] The three rows at the bottom of Table 5 with only two checkmarks are ambiguous; please add a legend or explicit column labels so the reader can tell which modules are active in each row.
[Section 3.2.2, Eq. (5)] The global context vector c is described as 'learnable,' but Eq. (5) computes it by pooling the auxiliary vectors; please clarify which parameters, if any, are learned in this step.
[Section 4.3.1] The text after the ATE equation contains a formatting artifact ('ATEisreflectiveoftheaveragedeviation'); the equation and surrounding text should be cleaned up.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the pipeline is trained and evaluated on external public datasets against external baselines, and no derivation step reduces to its own inputs.

full rationale

The paper's central claim is an empirical comparison: a semantic-graph encoder using graph attention networks and a graph comparison module using a difference term are trained with binary cross-entropy on pair labels generated from ground-truth poses on SemanticKITTI and KITTI-360, then evaluated with max F1 on held-out sequences. The graph vector construction (Eqs. 1-6) is a learned, data-dependent embedding, not a hand-fitted parameter renamed as a prediction; the comparison module (Eq. 7) consumes the learned graph vectors and is itself trained to predict similarity. The claimed 13% and 19% improvements over SGPR are comparative results against an external method, with the caveat that baseline provenance and run-to-run variability are not fully documented, but that is an experimental-validity concern, not circularity. The paper's dependencies on SGPR, SimGNN, GAT, and self-attention are externally published results, not self-citations, and no 'uniqueness theorem' is imported from the authors' prior work to force the design. The only apparent self-citation, Prakhya et al. (2017) in the ablation discussion of geometric descriptors, is not load-bearing because that descriptor was considered and not adopted. Thus the derivation chain is self-contained against external benchmarks and no circular step can be exhibited from the paper's own equations or citations.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The central claim rests on a standard supervised learning setup: no new physical entities, no parameter-free derivation. The free parameters are hyperparameters with mild sensitivity.

free parameters (4)

num_nodes = 50
Maximum number of graph nodes; if more instances, randomly sample 50; if fewer, pad with zero nodes. Ablation shows performance stable for >=50 nodes, but it is a manually chosen cap.
kNN_k = 10
Number of nearest neighbors in GAT aggregation; authors report different k had negligible impact.
graph_vector_dim = 32
Dimension of graph embedding; authors state 16-64 dimensions gave no noticeable change.
semantic_registration_weights = 1.2 for traffic signs/poles/buildings, 0.8 otherwise
Hand-set weights in the registration cost function (Eq. 9); not learned from data.

assumptions (3)

domain assumption Semantic labels and instance segmentation are available for input scans.
The whole method operates on semantic graphs built from labeled instances; KITTI uses ground-truth labels or RangeNet++ predictions.
domain assumption Object instances, centroids, and bounding boxes constitute a sufficient representation of a place.
The graph encoder compresses the scene into these node features; the paper does not prove this representation is discriminative beyond empirical results.
standard math Standard GAT and self-attention formulations (Eqs. 1-6) are valid and implementable as described.
Equations follow Veličković et al., Vaswani et al., and SimGNN.

how reviews work

0 comments

Cite this review

Pith. "Pith review of LiDAR Loop Closure Detection using Semantic Graphs with Graph Attention Networks." pith.science (2026). https://pith.science/paper/5F2JMZXS

@misc{pith2026250119382,
  author       = {Pith},
  title        = {Pith review of: LiDAR Loop Closure Detection using Semantic Graphs with Graph Attention Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5F2JMZXS}},
  note         = {Machine review of arXiv:2501.19382}
}

read the original abstract

In this paper, we propose a novel loop closure detection algorithm that uses graph attention neural networks to encode semantic graphs to perform place recognition and then use semantic registration to estimate the 6 DoF relative pose constraint. Our place recognition algorithm has two key modules, namely, a semantic graph encoder module and a graph comparison module. The semantic graph encoder employs graph attention networks to efficiently encode spatial, semantic and geometric information from the semantic graph of the input point cloud. We then use self-attention mechanism in both node-embedding and graph-embedding steps to create distinctive graph vectors. The graph vectors of the current scan and a keyframe scan are then compared in the graph comparison module to identify a possible loop closure. Specifically, employing the difference of the two graph vectors showed a significant improvement in performance, as shown in ablation studies. Lastly, we implemented a semantic registration algorithm that takes in loop closure candidate scans and estimates the relative 6 DoF pose constraint for the LiDAR SLAM system. Extensive evaluation on public datasets shows that our model is more accurate and robust, achieving 13% improvement in maximum F1 score on the SemanticKITTI dataset, when compared to the baseline semantic graph algorithm. For the benefit of the community, we open-source the complete implementation of our proposed algorithm and custom implementation of semantic registration at https://github.com/crepuscularlight/SemanticLoopClosure

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 32 canonical work pages

[1]

In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 5297--5307

Arandjelovic R, Gronat P, Torii A, et al (2016) NetVLAD: CNN architecture for weakly supervised place recognition . In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 5297--5307

work page 2016
[2]

In: Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining

Bai Y, Ding H, Bian S, et al (2019) SimGNN: A Neural Network Approach to Fast Graph Similarity Computation . In: Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining. Association for Computing Machinery, New York, NY, USA, WSDM '19, p 384–392, doi:10.1145/3289600.3290967, ://doi.org/10.1145/3289600.3290967

arXiv 2019
[3]

In: Proc

Behley J, Garbade M, Milioto A, et al (2019) SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences . In: Proc. of the IEEE/CVF International Conf. on Computer Vision (ICCV)

work page 2019
[4]

In: 2013 IEEE International Conference on Robotics and Automation, pp 2677--2684, doi:10.1109/ICRA.2013.6630945

Bosse M, Zlot R (2013) Place recognition using keypoint voting in large 3d lidar datasets. In: 2013 IEEE International Conference on Robotics and Automation, pp 2677--2684, doi:10.1109/ICRA.2013.6630945

work page arXiv 2013
[5]

Brody S, Alon U, Yahav E (2022) How attentive are graph attention networks? In: International Conference on Learning Representations, ://openreview.net/forum?id=F72ximsx7C1

work page 2022
[6]

In: International Conference on Learning Representations (ICLR2014), CBLS, April 2014

Bruna J, Zaremba W, Szlam A, et al (2014) Spectral networks and locally connected networks on graphs. In: International Conference on Learning Representations (ICLR2014), CBLS, April 2014

work page 2014
[7]

IEEE Transactions on Robotics 38(4):2074--2093

Cattaneo D, Vaghi M, Valada A (2022) LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAM . IEEE Transactions on Robotics 38(4):2074--2093

work page 2022
[8]

In: Proceedings of the IEEE/RSJ Int

Chen X, Milioto A, Palazzolo E, et al (2019) SuMa++: Efficient LiDAR-based Semantic SLAM . In: Proceedings of the IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS)

work page 2019

Show all 42 references

[9]

In: IEEE International Conference on Robotics and Automation (ICRA)

Dub \'e R, Dugas D, Stumm E, et al (2017) SegMatch: Segment Based Place Recognition in 3D Point Clouds . In: IEEE International Conference on Robotics and Automation (ICRA)

2017
[10]

In: Robotics: Science and Systems XIV

Dub \' e R, Cramariuc A, Dugas D, et al (2018) SegMap : 3D Segment Mapping using Data-Driven Descriptors . In: Robotics: Science and Systems XIV . Robotics: Science and Systems Foundation, doi:10.15607/rss.2018.xiv.003

2018 doi
[11]

In: Conference on Computer Vision and Pattern Recognition (CVPR)

Geiger A, Lenz P, Urtasun R (2012) Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite . In: Conference on Computer Vision and Pattern Recognition (CVPR)

2012
[12]

In: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 231--237, doi:10.1109/IROS.2016.7759060

He L, Wang X, Zhang H (2016) M2DP: A Novel 3D Point Cloud Descriptor and Its Application in Loop Closure Detection . In: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 231--237, doi:10.1109/IROS.2016.7759060

2016
[13]

Hong Y, Zhen H, Chen P, et al (2023) 3D-LLM: Injecting the 3D World into Large Language Models . arXiv

2023
[14]

In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 4802--4809, doi:10.1109/IROS.2018.8593953

Kim G, Kim A (2018) Scan Context: Egocentric Spatial Descriptor for Place Recognition Within 3D Point Cloud Map . In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 4802--4809, doi:10.1109/IROS.2018.8593953

2018
[15]

IEEE Transactions on Robotics 38(3):1856--1874

Kim G, Choi S, Kim A (2021) Scan Context++: Structural Place Recognition Robust to Rotation and Lateral Variations in Urban Environments . IEEE Transactions on Robotics 38(3):1856--1874

2021
[16]

In: International Conference on Learning Representations, ://openreview.net/forum?id=SJU4ayYgl

Kipf TN, Welling M (2017) Semi-Supervised Classification with Graph Convolutional Networks . In: International Conference on Learning Representations, ://openreview.net/forum?id=SJU4ayYgl

2017
[17]

arXiv preprint arXiv:230402643

Kirillov A, Mintun E, Ravi N, et al (2023) Segment anything . arXiv preprint arXiv:230402643

2023
[18]

In: 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), pp 1789--1798, doi:10.1109/WACV48630.2021.00183

Komorowski J (2021) MinkLoc3D: Point Cloud Based Large-Scale Place Recognition . In: 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), pp 1789--1798, doi:10.1109/WACV48630.2021.00183

2021
[19]

In: 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 8216--8223, doi:10.1109/IROS45743.2020.9341060

Kong X, Yang X, Zhai G, et al (2020) Semantic Graph Based Place Recognition for 3D Point Clouds . In: 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 8216--8223, doi:10.1109/IROS45743.2020.9341060

2020
[20]

IEEE Transactions on Pattern Analysis and Machine Intelligence 45(6):6923–6939

Li G, M\" u ller M, Qian G, et al (2023) DeepGCNs: Making GCNs Go as Deep as CNNs . IEEE Transactions on Pattern Analysis and Machine Intelligence 45(6):6923–6939. doi:10.1109/TPAMI.2021.3074057, ://doi.org/10.1109/TPAMI.2021.3074057

2023
[22]

In: 2021 IEEE International Conference on Robotics and Automation (ICRA), IEEE, pp 7627--7634

Li L, Kong X, Zhao X, et al (2021 b ) SA-LOAM: Semantic-aided LiDAR SLAM with Loop Closure . In: 2021 IEEE International Conference on Robotics and Automation (ICRA), IEEE, pp 7627--7634

2021
[23]

doi:10.48550/ARXIV.2109.13410, ://arxiv.org/abs/2109.13410

Liao Y, Xie J, Geiger A (2021) Kitti-360: A novel dataset and benchmarks for urban scene understanding in 2d and 3d. doi:10.48550/ARXIV.2109.13410, ://arxiv.org/abs/2109.13410

2021 doi arXiv
[24]

In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 2831--2840

Liu Z, Zhou S, Suo C, et al (2019) LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment Analysis . In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 2831--2840

2019
[25]

In: International Conference on Learning Representations, ://openreview.net/forum?id=Bkg6RiCqY7

Loshchilov I, Hutter F (2019) Decoupled Weight Decay Regularization . In: International Conference on Learning Representations, ://openreview.net/forum?id=Bkg6RiCqY7

2019
[26]

IEEE Robotics and Automation Letters 7(3):6958--6965

Ma J, Zhang J, Xu J, et al (2022) Overlaptransformer: An efficient and yaw-angle-invariant transformer network for lidar-based place recognition. IEEE Robotics and Automation Letters 7(3):6958--6965. doi:10.1109/LRA.2022.3178797

2022
[27]

In: 2009 IEEE International Conference on Robotics and Automation, pp 23--28, doi:10.1109/ROBOT.2009.5152712

Magnusson M, Andreasson H, Nuchter A, et al (2009) Appearance-based Loop Detection from 3D Laser data Using the Normal Distributions Transform . In: 2009 IEEE International Conference on Robotics and Automation, pp 23--28, doi:10.1109/ROBOT.2009.5152712

2009
[28]

IEEE Robotics and Automation Letters 2(3):1472--1479

Prakhya SM, Lin J, Chandrasekhar V, et al (2017) 3DHoPD: A Fast Low-Dimensional 3-D Descriptor . IEEE Robotics and Automation Letters 2(3):1472--1479. doi:10.1109/LRA.2017.2667721

2017
[29]

In: 2019 16th International Conference on Machine Vision Applications (MVA), pp 1--6, doi:10.23919/MVA.2019.8758020

Prokhorov D, Zhukov D, Barinova O, et al (2019) Measuring Robustness of Visual SLAM . In: 2019 16th International Conference on Machine Vision Applications (MVA), pp 1--6, doi:10.23919/MVA.2019.8758020

2019
[30]

In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 652--660

Qi CR, Su H, Mo K, et al (2017) Pointnet: Deep Learning on Point Sets for 3d Classification and Segmentation . In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 652--660

2017
[31]

In: ICRA Workshop on Open Source Software

Quigley M, Conley K, Gerkey B, et al (2009) ROS: An Open-Source Robot Operating System . In: ICRA Workshop on Open Source Software

2009
[32]

In: International conference on machine learning, PMLR, pp 8748--8763

Radford A, Kim JW, Hallacy C, et al (2021) Learning Transferable Visual Models from Natural Language Supervision . In: International conference on machine learning, PMLR, pp 8748--8763

2021
[33]

In: 2009 IEEE International Conference on Robotics and Automation, pp 3212--3217, doi:10.1109/ROBOT.2009.5152473

Rusu RB, Blodow N, Beetz M (2009) Fast Point Feature Histograms (FPFH) for 3D Registration . In: 2009 IEEE International Conference on Robotics and Automation, pp 3212--3217, doi:10.1109/ROBOT.2009.5152473

2009
[34]

Computer Vision and Image Understanding 125:251--264

Salti S, Tombari F, Di Stefano L (2014) SHOT: Unique Signatures of Histograms for Surface and Texture Description . Computer Vision and Image Understanding 125:251--264. doi:https://doi.org/10.1016/j.cviu.2014.04.011, ://www.sciencedirect.com/science/article/pii/S1077314214000988

2014 doi
[35]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Uy MA, Lee GH (2018) PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition . In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

2018
[36]

In: Guyon I, Luxburg UV, Bengio S, et al (eds) Advances in Neural Information Processing Systems, vol 30

Vaswani A, Shazeer N, Parmar N, et al (2017) Attention is All you Need . In: Guyon I, Luxburg UV, Bengio S, et al (eds) Advances in Neural Information Processing Systems, vol 30. Curran Associates, Inc., ://proceedings.neurips.cc/paper_files/paper/2017/file/3f5ee243547dee91fbd...

2017
[37]

In: International Conference on Learning Representations, ://openreview.net/forum?id=rJXMpikCZ

Veličković P, Cucurull G, Casanova A, et al (2018) Graph attention networks. In: International Conference on Learning Representations, ://openreview.net/forum?id=rJXMpikCZ

2018
[38]

In: 2020 IEEE International Conference on Robotics and Automation ( ICRA )

Wang H, Wang C, Xie L (2020) Intensity Scan Context: Coding Intensity and Geometry Relations for Loop Closure Detection . In: 2020 IEEE International Conference on Robotics and Automation ( ICRA ). IEEE , doi:10.1109/icra40945.2020.9196764

2020
[39]

In: 2021 IEEE / RSJ International Conference on Intelligent Robots and Systems ( IROS )

Wang H, Wang C, Chen CL, et al (2021) F- LOAM : Fast LiDAR odometry and mapping. In: 2021 IEEE / RSJ International Conference on Intelligent Robots and Systems ( IROS ). IEEE , doi:10.1109/iros51168.2021.9636655

2021
[40]

ACM Transactions on Graphics (tog) 38(5):1--12

Wang Y, Sun Y, Liu Z, et al (2019) Dynamic Graph CNN for Learning on Point Clouds . ACM Transactions on Graphics (tog) 38(5):1--12

2019
[41]

In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp 1912--1920, doi:10.1109/CVPR.2015.7298801

Wu Z, Song S, Khosla A, et al (2015) 3D ShapeNets: A Deep Representation for Volumetric Shapes . In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp 1912--1920, doi:10.1109/CVPR.2015.7298801

2015
[42]

, " * write output.state after.block = add.period write newline

ENTRY address archive author booktitle chapter doi edition editor eid eprint howpublished institution journal key keywords month note number organization pages publisher school series title type url volume year archivePrefix primaryClass adsurl adsnote version label extra.labe...
[43]

write newline

" write newline "" before.all 'output.state := FUNCTION add.period duplicate empty 'skip "." * add.blank if FUNCTION if.digit duplicate "0" = swap duplicate "1" = swap duplicate "2" = swap duplicate "3" = swap duplicate "4" = swap duplicate "5" = swap duplicate "6" = swap dupl...

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.

[1] [1]

In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 5297--5307

Arandjelovic R, Gronat P, Torii A, et al (2016) NetVLAD: CNN architecture for weakly supervised place recognition . In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 5297--5307

work page 2016

[2] [2]

In: Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining

Bai Y, Ding H, Bian S, et al (2019) SimGNN: A Neural Network Approach to Fast Graph Similarity Computation . In: Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining. Association for Computing Machinery, New York, NY, USA, WSDM '19, p 384–392, doi:10.1145/3289600.3290967, ://doi.org/10.1145/3289600.3290967

arXiv 2019

[3] [3]

In: Proc

Behley J, Garbade M, Milioto A, et al (2019) SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences . In: Proc. of the IEEE/CVF International Conf. on Computer Vision (ICCV)

work page 2019

[4] [4]

In: 2013 IEEE International Conference on Robotics and Automation, pp 2677--2684, doi:10.1109/ICRA.2013.6630945

Bosse M, Zlot R (2013) Place recognition using keypoint voting in large 3d lidar datasets. In: 2013 IEEE International Conference on Robotics and Automation, pp 2677--2684, doi:10.1109/ICRA.2013.6630945

work page arXiv 2013

[5] [5]

Brody S, Alon U, Yahav E (2022) How attentive are graph attention networks? In: International Conference on Learning Representations, ://openreview.net/forum?id=F72ximsx7C1

work page 2022

[6] [6]

In: International Conference on Learning Representations (ICLR2014), CBLS, April 2014

Bruna J, Zaremba W, Szlam A, et al (2014) Spectral networks and locally connected networks on graphs. In: International Conference on Learning Representations (ICLR2014), CBLS, April 2014

work page 2014

[7] [7]

IEEE Transactions on Robotics 38(4):2074--2093

Cattaneo D, Vaghi M, Valada A (2022) LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAM . IEEE Transactions on Robotics 38(4):2074--2093

work page 2022

[8] [8]

In: Proceedings of the IEEE/RSJ Int

Chen X, Milioto A, Palazzolo E, et al (2019) SuMa++: Efficient LiDAR-based Semantic SLAM . In: Proceedings of the IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS)

work page 2019