|
|
|
| Traffic Scene Semantic Segmentation Algorithm with Knowledge Distillation of Multi-level Features Guided by Boundary Perception |
| XIE Xinlin1,2, DUAN Zeyun1,2, LUO Chenyan1,2, XIE Gang1,2 |
1. School of Electronic Information Engineering, Taiyuan University of Science and Technology, Taiyuan 030024;
2. Shanxi Key Laboratory of Advanced Control and Equipment Intelligence, Taiyuan University of Science and Technology, Taiyuan 030024 |
|
|
|
|
Abstract To solve the problems of object detail information loss and large model parameters in traffic scenes, a traffic scene semantic segmentation algorithm with knowledge distillation of multi-level features guided by boundary perception is proposed. The proposed algorithm can smooth the object segmentation boundaries with fewer parameters. First, the adaptive fusing multi-level feature module is constructed to integrate the multi-level features of deep semantic information and shallow spatial information. The object boundary information and object subject information are highlighted selectively. Second, an interactive attention fusion module is proposed to model the long-range dependencies in spatial and channel dimensions, enhancing the information interaction capabilities between different dimensions. Finally, a boundary loss function based on candidate boundaries is proposed to construct a boundary knowledge distillation network based on detail awareness and transfer boundary information from complex teacher networks. Experiments on the traffic scene datasets Cityscapes and CamVid demonstrate that the proposed algorithm achieves a lightweight model while gaining positive segmentation performance, maintaining significant advantages in dealing with small and slender objects.
|
|
Received: 31 July 2024
|
|
|
| Fund:National Natural Science Foundation of China(No.62006169), Key Research and Development Program Fund of Shanxi Province(No.202202010101005), General Program of Basic Research Plan of Shanxi Province(No.202303021221141), Taiyuan City Key Core Technology Research and Development "List and Commander-in-Chief" Project(No.2024TYJB0137) |
|
Corresponding Authors:
XIE Xinlin, Ph.D., associate professor. His research interests include image semantic segmentation and deep learning.
|
| About author:: DUAN Zeyun, Master student. Her research interests include deep learning and image semantic segmentation. LUO Chenyan, Master student. Her research interests include deep learning and image semantic segmentation. XIE Gang, Ph.D., professor. His research interests include advanced control, machine vision and fault diagnosis. |
|
|
|
[1] PAN H H, HONG Y D, SUN W C, et al. Deep Dual-Resolution Networks for Real-Time and Accurate Semantic Segmentation of Tra-ffic Scenes. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(3): 3448-3460.
[2] HU X G, JING L Y, SEHAR U.Joint Pyramid Attention Network for Real-Time Semantic Segmentation of Urban Scenes. Applied Intelligence, 2022, 52(1): 580-594.
[3] 蒋斌,涂文轩,杨超,等.基于DenseNet的复杂交通场景语义分割方法.模式识别与人工智能, 2019, 32(5): 472-480.
(JIANG B, TU W X, YANG C, et al. Semantic Segmentation Method for Complex Traffic Scene Based on DenseNet. Pattern Re-cognition and Artificial Intelligence, 2019, 32(5): 472-480.)
[4] XIAO X Y, ZHAO Y Q, ZHANG F, et al. BASeg: Boundary Aware Semantic Segmentation for Autonomous Driving. Neural Networks, 2023, 157: 460-470.
[5] GAO G W, XU G A, LI J C.FBSNet: A Fast Bilateral Symmetrical Network for Real-Time Semantic Segmentation. IEEE Transactions on Multimedia, 2023, 25: 3273-3283.
[6] 张墺琦,亢宇鑫,武卓越,等.基于多尺度特征和注意力机制的肝脏组织病理图像语义分割网络.模式识别与人工智能, 2021, 34(4): 375-384.
(ZHANG A Q, KANG Y X, WU Z Y, et al. Semantic Segmentation Network of Pathological Images of Liver Tissue Based on the Multi-scale Feature and Attention Mechanism. Pattern Recognition and Artificial Intelligence, 2021, 34(4): 375-384.)
[7] FU J, LIU J, TIAN H J, et al. Dual Attention Network for Scene Segmentation//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 3141-3149.
[8] DAI Y M, GIESEKE F, OEHMCKE S, et al. Attentional Feature Fusion//Proc of the IEEE Winter Conference on Applications of Computer Vision. Washington, USA: IEEE, 2021: 3559-3568.
[9] ZHOU Z, ZHOU Y, WANG D L, et al. Self-Attention Feature Fusion Network for Semantic Segmentation. Neurocomputing, 2021, 453: 50-59.
[10] YANG C G, ZHOU H L, AN Z L, et al. Cross-Image Relational Knowledge Distillation for Semantic Segmentation//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2022: 12309-12318.
[11] JI D Y, WANG H R, TAO M Y.Structural and Statistical Texture Knowledge Distillation for Semantic Segmentation//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2022: 16855-16864.
[12] AN S M, LIAO Q M, LU Z Q, et al. Efficient Semantic Segmentation via Self-Attention and Self-Distillation. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(9): 15256-15266.
[13] HAN H Y, CHEN Y C, HSIAO P Y, et al. Using Channel-Wise Attention for Deep CNN Based Real-Time Semantic Segmentation with Class-Aware Edge Information. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(2): 1041-1051.
[14] GOU J P, YU B S, MAYBANK S J, et al. Knowledge Distillation: A Survey. International Journal of Computer Vision, 2021, 129(6): 1789-1819.
[15] HINTON G, ORIOL V, DEAN J.Distilling the Knowledge in a Neural Network[C/OL].[2024-06-20].https://arxiv.org/pdf/1503.02531.
[16] ZHANG Y, XIANG T, HOSPEDALES T M, et al. Deep Mutual Learning//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 4320-4328.
[17] ZHANG H R, HU Z Z, QIN W, et al. Adversarial Co-distillation Learning for Image Recognition. Pattern Recognition, 2021, 111. DOI: 10.1016/j.patcog.2020.107659.
[18] HOU Y N, MA Z, LIU C X, et al. Learning Lightweight Lane De-tection CNNs by Self-Attention Distillation//Proc of the IEEE/CVF International Conference on Computer Vision. Washington, USA: IEEE, 2019: 1013-1021.
[19] LIU Y A, ZHANG W, WANG J.Adaptive Multi-teacher Multi-level Knowledge Distillation. Neurocomputing, 2020, 415: 106-113.
[20] WANG Y K, ZHOU W, JIANG T.Intra-class Feature Variation Distillation for Semantic Segmentation//Proc of the 16th European Conference on Computer Vision. Berlin, Germany: Springer, 2020: 346-362.
[21] FENG Y C, SUN X, DIAO W H, et al. Double Similarity Disti-llation for Semantic Image Segmentation. IEEE Transactions on Image Processing, 2021, 30: 5363-5376.
[22] SHU C Y, LIU Y F, GAO J F.Channel-Wise Knowledge Disti-llation for Dense Prediction//Proc of the IEEE/CVF International Conference on Computer Vision. Washington, USA: IEEE, 2021: 5291-5300.
[23] DING X H, ZHANG X Y, MA N N, et al. RepVGG: Making VGG-Style ConvNets Great Again//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2021: 13728-13737.
[24] HE K M, ZHANG X Y, REN S Q, et al. Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification//Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2015: 1026-1034.
[25] YU C Q, WANG J B, PENG C,et al. BiSeNet: Bilateral Segmentation Network for Real-Time Semantic Segmentation//Proc of the European Conference on Computer Vision. Berlin, Germany: Sprin-ger, 2018: 334-349.
[26] 谢新林,罗臣彦,续欣莹,等.双注意力引导的跨层优化交通场景语义分割.交通运输系统工程与信息, 2023, 23(1): 236-244.
(XIE X L, LUO C Y, XU X Y,et al. Dual Attention Guided Cross-Layer Optimized Traffic Scene Semantic Segmentation. Journal of Transportation Systems Engineering and Information Techno-logy, 2023, 23(1): 236-244.)
[27] BADRINARAYANAN V, KENDALL A, CIPOLLA R.SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.
[28] ZHAO H S, SHI J P, QI X J, et al. Pyramid Scene Parsing Network//Proc of the IEEE Conference on Computer Vision and Pa-ttern Recognition. Washington, USA: IEEE, 2017: 6230-6239.
[29] ZHAO H S, QI X J, SHEN X Y,et al. ICNet for Real-Time Semantic Segmentation on High-Resolution Images//Proc of the European Conference on Computer Vision. Berlin, Germany: Sprin-ger, 2018: 418-434.
[30] ORSIC M, KRESO I, BEVANDIC P, et al. In Defense of Pre-trained ImageNet Architectures for Real-Time Semantic Segmentation of Road-Driving Images//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 12599-12608.
[31] CHEN L C, ZHU Y K, PAPANDREOU G, et al. Encoder-Deco-der with Atrous Separable Convolution for Semantic Image Segmentation//Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2018: 833-851.
[32] YANG L, BAI Y W, REN F L, et al. LCFNets: Compensation Strategy for Real-Time Semantic Segmentation of Autonomous Dri-ving. IEEE Transactions on Intelligent Vehicles, 2024, 9(4): 4715-4729.
[33] KARINE A, NAPOLÉON T, JRIDI M. Channel-Spatial Knowle-dge Distillation for Efficient Semantic Segmentation. Pattern Re-cognition Letters, 2024, 180: 48-54.
[34] WU Y, JIANG J Y, HUANG Z M, et al. FPANet: Feature Pyramid Aggregation Network for Real-Time Semantic Segmentation. Applied Intelligence, 2022, 52(3): 3319-3336. |
|
|
|
2024, Vol. 37 
