Abstract
As semantic segmentation models can accurately distinguish object category and contour from the background of images, its application in autonomous driving has been widely studied in recent years. As one of the most important perception modules in autonomous driving, many modern lane line detection models adopt the method of semantic segmentation or instance segmentation. Compared to other objects in traffic scenes, lane lines have a linear structure characteristic of regular shape. Based on that, we design a semantic segmentation model to take advantage of the special structure of lane lines. Combining VH-stage’s two-branches horizontal and vertical one-dimensional convolution with SCNN’s spatial convolution, we propose a multi spatial convolution block (MSCB). We evaluate our method on PSV and TSD datasets and find MSCB substantially improves the accuracy of lane lines semantic segmentation by up to 4%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Garcia-Garcia, A., Orts-Escolano,. S, Oprea, S., et al: A review on deep learning techniques applied to semantic segmentation. arXiv preprint arXiv:1704.06857 (2017)
LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436–444 (2015)
Wu, Y., Yang, T., Zhao, J., et al.: VH-HFCN based parking slot and lane markings segmentation on panoramic surround view. In: 2018 IEEE Intelligent Vehicles Symposium (IV), pp. 1767–1772. IEEE (2018)
Pan, X., Shi, J., Luo, P., et al.: Spatial as deep: spatial CNN for traffic scene understanding. arXiv preprint arXiv:1712.06080 (2017)
Jiang, W., Wu, Y., Guan, L., Zhao, J.: DFNet: semantic segmentation on panoramic images with dynamic loss weights and residual fusion block. In: 2019 International Conference on Robotics and Automation (ICRA), pp. 5887–5892. IEEE (2019)
Chiu, K.Y., Lin, S.F.: Lane detection using color-based segmentation. In: Intelligent Vehicles Symposium, pp. 706–711. IEEE (2005)
Loose, H., Franke, U., Stiller, C.: Kalman particle filter for lane recognition on rural roads. In: Intelligent Vehicles Symposium, pp. 60–65 IEEE (2009)
Seungwoo, Y., Heeseok, L., Heesoo, M., et al.: End-to-end lane marker detection via row-wise classification. arXiv preprint arXiv: 2005.08630 (2020)
Lee, S., Kim, J., Yoon, J.S., et al.: VPGNet: vanishing point guided network for lane and road marking detection and recognition. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1965–1973. IEEE (2017)
Lo, S.Y., Huang, H.M., Chan, S.W., et al.: Multi-class lane semantic segmentation using efficient convolutional networks. arXiv preprint arXiv:1907.09438 (2019)
Neven, D., Brabandere, B., Georgoulis, S., et al.: Towards end-to-end lane detection: an instance segmentation approach. In: 2018 IEEE Intelligent Vehicles Symposium (IV). pp. 286–291. IEEE (2018)
Chang, D., Chirakkal, V., Goswami, S., et al.: Multi-lane detection using instance segmentation and attentive voting. arXiv preprint arXiv:2001.00236 (2020)
Yu, Z., Ren, X., Huang, Y., et al.: Detecting lane and road markings at a distance with perspective transformer layers. arXiv preprint arXiv: 2003.08550 (2020)
Li, J., Mei, X., Prokhorov, D., et al.: Deep neural network for structural prediction and lane detection in traffic scene. In: IEEE Transactions on Neural Networks & Learning Systems, vol. 28, no. 3, pp. 690–703. IEEE (2017)
Institute of artificial intelligence and robotics, XJTU. TSD-max: Real traffic scene data set. http://trafficdata.xjtu.edu.cn/. Accessed 1 Mar 2021
Zhao, H., Shi, J., Qi, X., et al.: Pyramid scene parsing network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2881–2890. IEEE (2017)
Acknowledgments
This work was supported by the National Natural Science Foundation of China (No.U19A2069).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Wu, Y., Liu, F., Jiang, W., Yang, X. (2021). Multi Spatial Convolution Block for Lane Lines Semantic Segmentation. In: Huang, DS., Jo, KH., Li, J., Gribova, V., Hussain, A. (eds) Intelligent Computing Theories and Application. ICIC 2021. Lecture Notes in Computer Science(), vol 12837. Springer, Cham. https://doi.org/10.1007/978-3-030-84529-2_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-84529-2_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-84528-5
Online ISBN: 978-3-030-84529-2
eBook Packages: Computer ScienceComputer Science (R0)