Construction of a Dense Mapping System for Stereo Vision SLAM Based on Point-Line Fusion
Main Article Content
Keywords
visual SLAM, point-line fusion, wireframe parsing, desnse mapping
Abstract
To address issues such as insufficient feature extraction, inaccurate localization, and sparse point clouds encountered by visual SLAM systems in low-texture, dynamic environments, this paper proposes improvements to PLCD-SLAM. It introduces dual hardware-software constraints for depth estimation and presents a novel dense mapping architecture for stereo visual SLAM based on point-line fusion. Leveraging the robust Superpoint feature extractor and the strong structural capabilities of end-to-end wireframe parsing, the system compensates for feature information loss and enhances pose estimation accuracy. During depth computation, stereo matching combined with dual soft-hard constraints optimizes depth value precision, enabling high-accuracy dense mapping. To validate the improved algorithm's performance, comparative experiments were conducted on the KITTI, EUROC, and TUM (converted to stereo) public datasets. Results demonstrate that the proposed method achieves significantly lower absolute trajectory error (RMSE) than mainstream SLAM algorithms. It substantially enhances the system's robustness and positioning accuracy in dynamic and complex environments while producing high-quality dense maps.
References
- [1] Zhang C, Huang T, Zhang R, Yi X. PLD-SLAM: A New RGB-D SLAM Method with Point and Line Features for Indoor Dynamic Scene. ISPRS International Journal of Geo-Information. 2021; 10(3):163.
- [2] C. Forster, M. Pizzoli and D. Scaramuzza, SVO: Fast semi-direct monocular visual odometry, IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, China, 2014, 15-22.
- [3] Zhang J, Singh S. LOAM: Lidar odometry and mapping in real-time[C]//Robotics: Science and systems. 2014, 2(9): 1-9.
- [4] Gomez-Ojeda R, Moreno F A, Zuniga-Noël D, et al. PL-SLAM: A stereo SLAM system through the combination of points and line segments[J]. IEEE Transactions on Robotics, 2019, 35(3): 734-746.
- [5] Xu, Kuan et al. AirSLAM: An Efficient and Illumination-Robust Point-Line Visual SLAM System. IEEE Transactions on Robotics 2024, (41): 1673-1692.
- [6] DeTone D, Malisiewicz T, Rabinovich A. Superpoint: Self-supervised interest point detection and description[C]//Proceedings of the IEEE conference on computer vision and pattern recognition workshops. 2018: 224-236.
- [7] HANG Chenyang, YANG Jian. A Visual SLAM Method Coupled with Adaptive Point-Line Features and IMU[J]. Geomatics and Information Science of Wuhan University, 2025, 50(10): 2048-2063.
- [8] Gallagher L, Kumar V R, Yogamani S, et al. A hybrid sparse-dense monocular slam system for autonomous driving[C]//2021 European Conference on Mobile Robots (ECMR). IEEE, 2021: 1-8.
- [9] Wimbauer F, Yang N, Von Stumberg L, et al. MonoRec: Semi-supervised dense reconstruction in dynamic environments from a single moving camera[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 6112-6122.
- [10] Liu Y, Dong S, Wang S, et al. Slam3r: Real-time dense scene reconstruction from monocular rgb videos[C]//Proceedings of the Computer Vision and Pattern Recognition Conference. 2025: 16651-16662.
- [11] Yan C, Qu D, Xu D, et al. Gs-slam: Dense visual slam with 3d gaussian splatting[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2024: 19595-19604.
- [12] Engel J, Schöps T, Cremers D. LSD-SLAM: Large-scale direct monocular SLAM[C]//European conference on computer vision. Cham: Springer International Publishing, 2014: 834-849.
- [13] Wang R, Schworer M, Cremers D. Stereo DSO: Large-scale direct sparse visual odometry with stereo cameras[C]//Proceedings of the IEEE international conference on computer vision. 2017: 3903-3911.
- [14] Teed Z, Deng J. Droid-slam: Deep visual slam for monocular, stereo, and rgb-d cameras[J]. Advances in neural information processing systems, 2021, 34: 16558-16569.
- [15] Mur-Artal, Raul and Juan D. Tardós. ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras. IEEE Transactions on Robotics, 2016, (33):1255-1262.
- [16] Xue et al. Construction of Dense Stereo Maps for Orchards Based on Adaptive Threshold ORB Feature Extraction [J]. Transactions of the Chinese Society of Agricultural Machinery, 2024.
- [17] Wang et al. Research on Cross-Attention-Driven Dense Mapping Algorithm for Outdoor Stereo Vision SLAM [J]. Journal of Chongqing Technology and Forestry University (Natural Science), 2025.
- [18] Hu Y, Zhen W, Scherer S. Deep-learning assisted high-resolution binocular stereo depth reconstruction[C]//2020 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2020: 8637-8643.
- [19] Koestler L, Yang N, Zeller N, et al. Tandem: Tracking and dense mapping in real-time using deep multi-view stereo[C]//Conference on Robot Learning. PMLR, 2022: 34-45.
- [20] Burri M, Nikolic J, Gohl P, et al. The EuRoC micro aerial vehicle datasets[J]. The International Journal of Robotics Research, 2016, 35(10): 1157-1163.
- [21] Geiger A, Lenz P, Stiller C, et al. Vision meets robotics: The kitti dataset[J]. The international journal of robotics research, 2013, 32(11): 1231-1237.
- [22] Schubert D, Goll T, Demmel N, et al. The TUM VI benchmark for evaluating visual-inertial odometry[C]//2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2018: 1680-1687.
- [23] Yang N, Stumberg L, Wang R, et al. D3vo: Deep depth, deep pose and deep uncertainty for monocular visual odometry[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 1281-1292.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.