SeqLPD: Sequence Matching Enhanced Loop-Closure Detection Based on Large-Scale Point Cloud Description for Self-Driving Vehicles

Liu, Zhe; Suo, Chuanzhe; Zhou, Shunbo; Xu, Fan; Wei, Huanshu; Chen, Wen; Wang, Hesheng; Liang, Xinwu; Liu, Yunhui

doi:10.1109/iros40897.2019.8967875

Cited by 43 publications

(19 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But the previous work lacks an essential component of latent space representation, which is the reconstruction. Comparison with our previous work [12], [54], we significantly improve the discrimination performance of LPD-Net and study the reconstruction of the large-scale scene in this paper for the first time.…”

Section: B Place Recognitionmentioning

confidence: 79%

“…Our proposed LPD-net [12], as an encoder for latent space representation, has been optimized to meet the needs of the large-scale point cloud, which strengthened the geometric structure and neighborhood relationship to extracted the discriminative global descriptor. SepLPD [54] has verified its practicability and effectiveness in autonomous driving applications. But the previous work lacks an essential component of latent space representation, which is the reconstruction.…”

Section: B Place Recognitionmentioning

confidence: 91%

“…Various features in [12], [38], [54] can be extracted from the local structure, and we chose to describe the local structure of each point p i with features based on 3D eigenvalues, 2D eigenvalues, and 1D Z-direction distribution.…”

Section: ) Adaptive Neighborhood Feature Extractionmentioning

confidence: 99%

See 2 more Smart Citations

LPD-AE: Latent Space Representation of Large-Scale 3D Point Cloud

Suo

Liu

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

The effective latent space representation of point cloud provides a foremost and fundamental manner that can be used for challenging tasks, including point cloud based place recognition and reconstruction, especially in large-scale dynamic environments. In this paper, we present a novel deep neural network, LPD-AE(Large-scale Place Description AutoEncoder Network), to obtain meaningful local and contextual features for the generation of latent space from 3D point cloud directly. The encoder network constructs the discriminative global descriptors to realize high accuracy and robust place recognition, which contributed by extracting the local neighbor geometric features and aggregating neighborhood relationships both in feature space and physical space. The decoder network performs hierarchical reconstruction on coarse key points and ultimately produce dense point clouds, which shows that it is capable of reconstructing a full point cloud frame from a single compact but high dimensional descriptor. Our proposed network demonstrates performance that is comparable to the state-of-the-art approaches. With the benefit of the LPD-AE, many computationally complex tasks that rely directly on point clouds can be effortlessly conducted on latent space with lower memory costs, such as relocalization, loop closure detection, and map compression reconstruction. Comprehensive validations on Oxford RobotCar dataset, KITTI dataset, and our freshly collected dataset, which contains multiple trials of repeated routes in different weather and at different times, manifest its potency for real robotic and self-driving implementation.

show abstract

Section: B Place Recognitionmentioning

confidence: 79%

Section: B Place Recognitionmentioning

confidence: 91%

See 1 more Smart Citation

LPD-AE: Latent Space Representation of Large-Scale 3D Point Cloud

Suo

Liu

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…1, and similar demonstration is also given in [3]. Although some vision-based loop closure (or place recognition) algorithms [4], [5] were proposed to eliminate the drifts for terrestrial environments, they cannot directly handle the large-scale environment with real-world underwater data. In addition, due to the nature of underwater environments, changing viewpoints, blur motions, textureless images, and fast-moving objects further increase the difficulty of solving the problem.…”

Section: Introductionmentioning

confidence: 89%

Robust AUV Visual Loop-Closure Detection Based on Variational Autoencoder Network

Wang

et al. 2022

IEEE Trans. Ind. Inf.

View full text Add to dashboard Cite

The visual loop closure detection for Autonomous Underwater Vehicles (AUVs) is a key component to reduce the drift error accumulated in simultaneous localization and mapping tasks. However, due to viewpoint changes, textureless images, and fast-moving objects, the loop closure detection in dramatically changing underwater environments remains a challenging problem to traditional geometric methods. Inspired by strong feature learning ability of deep neural networks, we propose an underwater loop closure detection method based on a variational auto-encoder network in this paper. Our proposed method can learn effective image representations to deal with the challenges caused by dynamic underwater environments. Specifically, the proposed network is an unsupervised method, which avoids the difficulty and cost of labeling a great quantity of underwater data. Also included is a semantic object segmentation module, which is utilized to segment the underwater environments and assign weights to objects in order to alleviate the impact of fastmoving objects. Furthermore, an underwater image description scheme is used to enable efficient access to geometric and objectlevel semantic information, which helps to build a robust and real-time system in dramatically changing underwater scenarios. Finally, we test the proposed system under complex underwater environments and get a recall rate of 92.31% in the tested environments.

show abstract

“…However, PCAN may ignore the prior information of input data and it leads to high cost in the proposed attention module. SeqLPD [4] and LPD-AE [47] utilize LPD-Net as a place recognition module to implement environment construction. Moreover, [15] projects input point cloud into cylindrical coordinates and converts 3D point cloud to 2D image for place recognition.…”

Section: Related Workmentioning

confidence: 99%

WADE-Net: Weighted Aggregation with Density Estimation for Point Cloud Place Recognition

Liu¹,

Wang²,

Peng³

et al. 2021

APM

View full text Add to dashboard Cite

Point cloud based place recognition plays an important role in mobile robotics. In this paper, we propose a weighted aggregation method from structure information adaptively for point cloud place recognition. Firstly, to preserve the prior distributions and local geometric structures, we fuse learned hidden features with handcrafted features in the beginning. Secondly, we further extract and aggregate adaptively weighted features concerning density and relative spatial information from these fused features, named Weighted Aggregation with Density Estimation (WADE) module. Then, we conduct the WADE block iteratively to group the latent manifold structures. Finally, comparison results on two public datasets Oxford Robotcar and KITTI show that the proposed approach exceeds the comparison approaches on recall rate averagely 7% -8%.

show abstract

SeqLPD: Sequence Matching Enhanced Loop-Closure Detection Based on Large-Scale Point Cloud Description for Self-Driving Vehicles

Cited by 43 publications

References 20 publications

LPD-AE: Latent Space Representation of Large-Scale 3D Point Cloud

LPD-AE: Latent Space Representation of Large-Scale 3D Point Cloud

Robust AUV Visual Loop-Closure Detection Based on Variational Autoencoder Network

WADE-Net: Weighted Aggregation with Density Estimation for Point Cloud Place Recognition

Contact Info

Product

Resources

About