相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Proceedings Paper
Robotics
Tiago Barros et al.
Summary: Place recognition is crucial for SLAM and localization systems in autonomous navigation. With the advancement of deep learning in extracting information from 3D LiDARs, this modality has also improved the performance in place recognition. In this work, a novel deep learning network based on 3D LiDAR is proposed, utilizing a self-attention mechanism to extract invariant descriptors and enhance performance.
ROBOT2022: FIFTH IBERIAN ROBOTICS CONFERENCE: ADVANCES IN ROBOTICS, VOL 1
(2023)
Article
Computer Science, Artificial Intelligence
Yiyi Liao et al.
Summary: In recent years, the subfields of artificial intelligence, such as computer vision, graphics, and robotics, have developed independently. However, there is now a realization that progress towards robust intelligent systems requires collaboration across these fields. The development of KITTI-360 dataset aims to facilitate research at the intersection of vision, graphics, and robotics, and address challenges like fully autonomous self-driving systems.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Robotics
Lin Li et al.
Summary: LiDAR-based place recognition is a fundamental capability of robots, and this letter proposes a rotation invariant neural network structure for it. The network combines semantic and geometric features, design a rotation equivariant global descriptor, and uses a rotation invariant siamese neural network to predict descriptor pair similarity. The evaluations and tests on multiple datasets show that this approach can work stably in various scenarios and achieve state-of-the-art performance.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Computer Science, Artificial Intelligence
Xieyuanli Chen et al.
Summary: The paper introduces a modified Siamese network for estimating the similarity between pairs of LiDAR scans, which can be used for loop closing in SLAM and global localization in autonomous systems. The approach utilizes a deep neural network to exploit cues generated from LiDAR data, demonstrating superior performance and generalization in different environments. The method effectively detects loop closures and reliably localizes vehicles globally in urban environments using LiDAR data collected in various seasons.
Article
Robotics
Junyi Ma et al.
Summary: This article addresses the problem of place recognition based on 3D LiDAR scans and proposes a novel lightweight neural network approach. The author achieves fast execution using the range image representation of LiDAR sensors and designs a yaw-angle-invariant network architecture to improve place recognition performance. Experimental results demonstrate that the method performs well across different environments.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Robotics
Kenji Koide et al.
Summary: This study introduces an interactive graph SLAM framework with a 3D LIDAR, allowing users to correct 3D environmental maps interactively and achieve globally consistent maps through pose graph optimization and map correction constraints. The proposed semi-automatic loop closing and plane-based map correction techniques, along with a pose constraint update approach, improve mapping consistency and accuracy with minimal human effort, outperforming current automatic SLAM frameworks.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Robotics
Peng Yin et al.
Summary: The letter introduces FusionVLAD, a fusion-based network for real-time 3D place recognition, which encodes a multiview representation of sparse 3D point clouds. It consists of two parallel branches for orientation-invariant and translation-insensitive feature extraction, with a parallel fusion module to enhance the combination of region-wise feature connection between the two branches. Experiments show that FusionVLAD outperforms state-of-the-art methods in terms of accuracy and efficiency.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Le Hui et al.
Summary: This paper proposes a Pyramid Point Cloud Transformer Network (PPT-Net) to learn discriminative global descriptors from point clouds for efficient retrieval. By developing a pyramid point transformer module and a pyramid VLAD module, the method extracts discriminative local features and aggregates multi-scale feature maps to achieve state-of-the-art results in point cloud based place recognition task.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Yan Xia et al.
Summary: This paper introduces a novel self-attention and orientation encoding network (SOE-Net) for handling place recognition from point cloud data, achieving superior performance by fully exploring the relationship between points and incorporating long-range context into point-wise local descriptors. Additionally, a new loss function called HPHN quadruplet is proposed, which outperforms commonly used metric learning losses.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Jacek Komorowski Warsaw
Summary: The paper introduces a learning-based method, MinkLoc3D, for computing discriminative 3D point cloud descriptors for place recognition. The method utilizes sparse voxelized point cloud representation and sparse 3D convolutions to achieve improved performance, outperforming current state-of-the-art methods in standard benchmarks.
2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2021)
(2021)
Proceedings Paper
Automation & Control Systems
Peng Yin et al.
2020 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhe Liu et al.
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Mikaela Angelina Uy et al.
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2018)
Article
Computer Science, Artificial Intelligence
Christian Hane et al.
IMAGE AND VISION COMPUTING
(2017)
