Abstract
The development of Driverless Car, Unmanned Aerial Vehicle, Human–Computer Interaction and Artificial Intelligence has promoted the Internet of Things (IoT) industry, in which, Visual Simultaneous Localization and Mapping (VSLAM) is an important Localization and Mapping technique. Closed loop detection can alleviate the error accumulation during the operation of VSLAM. The traditional closed loop detection methods mostly rely on manually defined features, subjective and unstable, which are difficult to cope with complex and repetitive scenarios. Thus, triplet loss-based metric learning has been considered as a better solution for closed loop detection. In this paper, first, constructed Generator is applied to generate feature vector of hard negative sample. Second, triplet loss and generative loss have been applied to construct loss function. The keyframes are converted into feature vectors with well-trained model, evaluating the similarity of keyframes by calculating their distance of feature vectors, which is used to determine whether a closed loop is formed. Finally, TUM dataset is introduced to evaluate the Precision and Recall of the proposed metric learning. The well-trained model is applied to establish loop closing thread for VSLAM system. The experimental results illustrate the feasibility and effectiveness of the metric learning-based closed loop detection, which can be further applied to practical VSLAM systems.
Similar content being viewed by others
References
Ruan, F., et al.: Deep learning for real-time image steganalysis: a survey. J. Real Time Image Process. 17, 2 (2019)
Stasse, O.: SLAM and Vision-based Humanoid Navigation. A Reference, Humanoid Robotics (2019)
Lowry, S., et al.: Visual place recognition: a survey. IEEE Trans. Robotics 32(1), 1–19 (2016)
Fan, T., Wang, H., Rubenstein, M., Murphey, T.: CPL-SLAM: Efficient and Certifiably Correct Planar Graph-Based SLAM Using the Complex Number Representation. IEEE Trans. Robotics. (2020). https://doi.org/10.1109/TRO.2020.3006717
Gomez-Ojeda, R., Moreno, F., Zuñiga-Noël, D., Scaramuzza, D., Gonzalez-Jimenez, J.: PL-SLAM: a Stereo SLAM system through the combination of points and line segments. IEEE Trans. Robot. 35(3), 734–746 (2019)
Mangelson, J.G., Ghaffari, M., Vasudevan, R., Eustice, R.M.: Characterizing the Uncertainty of Jointly Distributed Poses in the Lie Algebra. IEEE Trans. Robot. (2020). https://doi.org/10.1109/TRO.2020.2994457
Wang, C., Zhai, Y., Xu, S., et al.: Multi-sensors based simultaneous mapping and global pose optimization. In: 2019 IEEE International Conference on Unmanned Systems (ICUS). IEEE (2020)
Memon, A.R., Wang, H., Hussain, A.: Loop closure detection using supervised and unsupervised deep neural networks for monocular SLAM systems. Robot. Autonom. Syst. 126, 103470 (2020)
Liu, R., Marakkalage, S.H., Padmal, M., et al.: Collaborative SLAM Based on WiFi fingerprint similarity and motion information. IEEE Internet Things J. 7(3), 1826–1840 (2020)
Filliat, D.: A visual bag of words method for interactive qualitative localization and mapping. In: 2007 IEEE international conference on robotics and automation (ICRA) (pp. 3921–3926), IEEE (2007)
Kwon, H., Yousef, K.M.A., Kak, A.C.: Building 3d visual maps of interior space with a new hierarchical sensor fusion architecture. Robot. Autonom. Syst. 61(8), 749–767 (2013)
Cummins, M., Newman, P.: FAB-MAP: probabilistic localization and mapping in the space of appearance. Int. J. Robot. Res. 27(6), 647-665 (2008)
Milford, M.J., Wyeth, G. F.: SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights. In: Proc. IEEE Int. Conf. Robot. Automat., pp. 1643–1649 (2012)
Maddern, W., Milford, M., Wyeth, G.: CAT-SLAM: Probabilistic localisation and mapping using a continuous appearance-based trajectory. Int. J. Robot. Res. 31(4), 429–451 (2012)
Pepperell, E., Corke, P.I., Milford, M.J.: All-environment visual place recognition with SMART. In: Proc. IEEE Int. Conf. Robot. Automat., pp. 1612–1618 (2014)
Dong, R., Wei, Z., Liu, C., Kan, J.: A novel loop closure detection method using line features. IEEE Access 7, 111245–111256 (2019)
Yang, Z., Pan, Y., Huan, R., Bao, Y.: Gridding place recognition for fast loop closure detection on mobile platforms. Electron. Lett. 55(17), 931–933 (2019)
Zhang, G., Yan, X., Ye, Y.: Loop closure detection via maximization of mutual information. IEEE Access 7, 124217–124232 (2019)
Kostavelis, I., Gasteratos, A.: Learning spatially semantic representations for cognitive robot navigation. Robot. Autonom. Syst. 61(12), 1460–1475 (2013)
Lepetit, V., Fua, P.: Keypoint recognition using randomized trees. IEEE Trans. Pattern Anal. Mach. Intell. 28(9), 1465–1479 (2006)
Chen, Z.T., Lam, O., Jacobson, A., et al.: Convolutional neural network-based place recognition[EB/OL]. (09-06) (2014)
Cummins, M., Newman, P.: FAB-MAP: probabilistic localization and mapping in the space of appearance. Int. J. Robot. Res. 27(6), 647-665 (2008)
Girshick, R., Donahue, J., Darrell, T., et al.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, USA: IEEE, pp. 580–587 (2014)
Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 17981828 (2013)
Wang, Z., Peng, Z., Guan, Y., Wu, L.: Manifold regularization graph structure auto-encoder to detect loop closure for visual SLAM. IEEE Access 7, 59524–59538 (2019)
Khaliq, A., Ehsan, S., Chen, Z., Milford, M., McDonald-Maier, K.: A Holistic visual place recognition approach using lightweight CNNs for significant ViewPoint and appearance changes. IEEE Trans. Robot. 36(2), 561–569 (2020)
Wang, S., Lv, X., Liu, X., Ye, D.: Compressed Holistic ConvNet Representations for Detecting Loop Closures in Dynamic Environments. IEEE Access 8, 60552–60574 (2020)
Hou, Y., Zhang, H., Zhou, S.L.: Convolutional neural networkbased image representation for visual loop closure detection. In: IEEE International Conference on Information and Automation. Piscataway, USA, IEEE, pp. 2238–2245 (2015)
Gao, X., Zhang, T.: Unsupervised learning to detect loops using deep neural networks for visual SLAM system. Autonom. Robots 41(1), 1–18 (2017)
Gao, X., Zhang, T.: Loop closure detection for visual SLAM systems using deep neural networks. In: 2015 34th Chinese Control Conference (CCC). IEEE (2015)
Sünderhauf, N., Dayoub, F., Shirazi, S., Upcroft, B., Milford, M.: On the performance of ConvNet features for place recognition. In: Proc. IEEE/RSJ Int. Conf. Intell. Robot. Syst., pp. 4297–4304 (2015)
Chen, W., Chen, X., Zhang, J., Huang, K.: A multi-task deep network for person re-identification. AAAI 1, 5 (2017)
Zheng, L., Yang, Y., Hauptmann. , A.G.: Person Reidentification: past, present and future. arXiv:1610.02984 (2016)
Geng, M., Wang, Y., Xiang, T., Tian, Y.: Deep transfer learning for person re-identification. arXiv:1611.05244 (2016)
Li, W., Zhu, X., Gong, S.: Person re-identification by deep joint learning of multi-loss classification. In IJCAI (2017)
Zheng, Z., Zheng, L., Yang, Y.: A Discriminatively Learned CNN Embedding for Person Re-identification. arXiv:1611.05666 (2016)
Schroff, F., Kalenichenko, D., Philbin, J.: FaceNet: A unified embedding for face recognition and clustering. In: CVPR (2015)
Shi, H., Yang, Y., Zhu, X., Liao, S., Lei, Z., Zheng, W., Li, S.Z.: Embedding Deep Metric for Person Re-identification: a study against Large Variations. In: ECCV (2016)
Faghri, F., Fleet, D.J., Kiros, J.R., et al.: VSE++: Improving visual-semantic embeddings with hard negatives (2017)
Duan, Y., Zheng, W., Lin, X., Lu, J., Zhou, J.: Deep adversarial metric learning. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE (2018)
Duan, Y., Lu, J., Zheng, W., Zhou, J.: Deep adversarial metric learning. IEEE Trans. Image Process. 29, 2037–2051 (2020)
Weinberger, K.Q.: Distance metric learning for large margin nearest neighbor classification. JMLR 2019, 10 (2009)
Mur-Artal, R., Tardos, J.D.: ORB-SLAM2: an open-source SLAM system for monocular, stereo and RGB-D Cameras. IEEE Trans. Robot. 33(5), 1255–1262 (2016)
https://www.robots.ox.ac.uk. Accessed 16 Jan 2021
https://www.robots.ox.ac.uk/mjc/Software.htm. Accessed 16 Jan 2021
https://github.com/rmsalinas/DBow3. Accessed 16 Jan 2021
Glover, A., Maddern, W., Warren, M., Reid, S., Milford, M., Wyeth, G.: OpenFABMAP: An open source toolbox for appearance-based loop closure detection.In: 2012 IEEE International Conference on Robotics and Automation, Saint Paul, MN, pp. 4730–4735 (2012)
Acknowledgements
This work was supported by the National Natural Science Foundation of China under the No. 61773282. The authors sincerely thank the anonymous reviewers for their valuable comments.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Chang, J., Dong, N., Li, D. et al. Metric learning with generator for closed loop detection in VSLAM. J Real-Time Image Proc 18, 1025–1036 (2021). https://doi.org/10.1007/s11554-020-01067-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11554-020-01067-7