Lane Segmentation to Optimize Traffic Flow
Abstract
Purpose. A comprehensive analysis of modern neural network architectures for lane segmentation aimed at determining their potential for optimizing traffic flows and evaluating their applicability in intelligent traffic management systems.
Method. A comparative analysis of five state-of-the-art deep learning architectures was conducted: LaneATT with an attention mechanism for keypoint detection; ERFNet and ENet as efficient real-time architectures; PINet for instance-based lane segmentation; and CondLaneNet with conditional lane shape generation. The study examines the specific characteristics of each architecture, their computational efficiency, and accuracy. The impact of segmentation quality on traffic flow optimization metrics is also analyzed.
Findings. The results show that CondLaneNet provides the highest segmentation accuracy and the best reconstruction of lane geometry under low traffic load and simple road topology, whereas LaneATT demonstrates slightly lower peak accuracy but exhibits smoother degradation in segmentation quality as traffic density increases. ENet and ERFNet provide a reasonable compromise between accuracy and computational complexity, making them suitable for real-time systems with limited resources. PINet, due to keypoint-based representations and clustering, demonstrates superior robustness to occlusion and complex road scenarios, maintaining a larger proportion of baseline quality under medium and high traffic conditions.
Practical implications. The findings support informed selection of the optimal architecture for specific application scenarios in intelligent transportation systems.
Paper type. Theoretical.
Downloads
References
Zheng, T., Huang, Y., Liu, Y., Tang, W., Yang, Z., Cai, D., & He, X. (2022). CLRNet: Cross layer refinement network for lane detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 898-907. https://doi.org/10.1109/CVPR52688.2022.00097
Liu, L., Chen, X., Zhu, S., & Tan, P. (2021). CondLaneNet: A top-to-down lane detection framework based on conditional convolution. Proceedings of the IEEE/CVF International Conference on Computer Vision, 3773-3782. https://doi.org/10.1109/ICCV48922.2021.00375
Ko, Y., Lee, Y., Azam, S., Munir, F., Jeon, M., & Pedrycz, W. (2022). Key points estimation and point instance segmentation approach for lane detection. IEEE Transactions on Intelligent Transportation Systems, 23(7), 8949-8958. https://doi.org/10.1109/TITS.2021.3088488
Tabelini, L., Berriel, R., Paixao, T. M., Badue, C., De Souza, A. F., & Oliveira-Santos, T. (2021). Keep your eyes on the lane: Real-time attention-guided lane detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 294-302. https://doi.org/10.1109/CVPR46437.2021.00036
Pan, X., Shi, J., Luo, P., Wang, X., & Tang, X. (2018). Spatial as deep: Spatial CNN for traffic scene understanding. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1), 7276-7283. https://doi.org/10.1609/aaai.v32i1.12301
Qin, Z., Wang, H., & Li, X. (2020). Ultra fast structure-aware deep lane detection. European Conference on Computer Vision (ECCV), 276-291. https://doi.org/10.1007/978-3-030-58586-0_17
Feng, Z., Guo, S., Tan, X., Xu, K., Wang, M., & Ma, L. (2022). Rethinking efficient lane detection via curve modeling. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 17062-17070. https://doi.org/10.1109/CVPR52688.2022.01655
Zheng, T., Fang, H., Zhang, Y., Tang, W., Yang, Z., Liu, H., & Cai, D. (2021). RESA: Recurrent feature-shift aggregator for lane detection. Proceedings of the AAAI Conference on Artificial Intelligence, 35(4), 3547-3554. https://doi.org/10.1609/aaai.v35i4.16469
Wang, J., Ma, Y., Huang, S., Hui, T., Wang, F., Qian, C., & Zhang, T. (2022). A keypoint-based global association network for lane detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1392-1401. https://doi.org/10.1109/CVPR52688.2022.00147
Liu, R., Yuan, Z., Liu, T., & Xiong, Z. (2021). End-to-end lane shape prediction with transformers. 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), 3693-3701. https://doi.org/10.1109/WACV48630.2021.00374
Xu, H., Wang, S., Cai, X., Zhang, W., Liang, X., & Li, Z. (2020). CurveLane-NAS: Unifying lane-sensitive architecture search and adaptive point blending. European Conference on Computer Vision (ECCV), 689-704. https://doi.org/10.48550/arXiv.2007.12147
Li, X., Li, J., Hu, X., & Yang, J. (2020). Line-CNN: End-to-end traffic line detection with line proposal unit. IEEE Transactions on Intelligent Transportation Systems, 21(1), 248-258. https://doi.org/10.1109/TITS.2019.2890870
Copyright (c) 2025 Liubomyr Drevych
This work is licensed under a Creative Commons Attribution 4.0 International License.
The authors agree with the following conditions:
1. Authors retain copyright and grant the journal right of first publication (Download agreement) with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
2. Authors have the right to complete individual additional agreements for the non-exclusive spreading of the journal’s published version of the work (for example, to post work in the electronic repository of the institution or to publish it as part of a monograph), with the reference to the first publication of the work in this journal.
3. Journal’s politics allows and encourages the placement on the Internet (for example, in the repositories of institutions, personal websites, SSRN, ResearchGate, MPRA, SSOAR, etc.) manuscript of the work by the authors, before and during the process of viewing it by this journal, because it can lead to a productive research discussion and positively affect the efficiency and dynamics of citing the published work (see The Effect of Open Access).
