LiteMSNet: a lightweight semantic segmentation network with multi-scale feature extraction for urban streetscape scenes

• Published in: The Visual computer Vol. 41; no. 4; pp. 2801 - 2815
• Main Authors: Li, Lirong, Ding, Jiang, Cui, Hao, Chen, Zhiqiang, Liao, Guisheng
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.03.2025; Springer Nature B.V
• Subjects: Accuracy; Artificial Intelligence; Computer Graphics; Computer Science; Datasets; Deep learning; Economic activity; Efficiency; Feature extraction; Image Processing and Computer Vision; Modules; Real time; Scene analysis; Semantic segmentation; Semantics; Dilated convolution; Semantic segmentation; Urban streetscape scene; Multi-scale features; Lightweight network
• ISSN: 0178-2789; 1432-2315
• DOI: 10.1007/s00371-024-03569-y

Semantic segmentation plays a pivotal role in computer scene understanding, but it typically requires a large amount of computing to achieve high performance. To achieve a balance between accuracy and complexity, we propose a lightweight semantic segmentation model, termed LiteMSNet (a Lightweight Semantic Segmentation Network with Multi-Scale Feature Extraction for urban streetscape scenes). In this model, we propose a novel Improved Feature Pyramid Network, which embeds a shuffle attention mechanism followed by a stacked Depth-wise Asymmetric Gating Module. Furthermore, a Multi-scale Dilation Pyramid Module is developed to expand the receptive field and capture multi-scale feature information. Finally, the proposed lightweight model integrates two loss mechanisms, the Cross-Entropy and the Dice Loss functions, which effectively mitigate the issue of data imbalance and gradient saturation. Numerical experimental results on the CamVid dataset demonstrate a remarkable mIoU measurement of 70.85% with less than 5M parameters, accompanied by a real-time inference speed of 66.1 FPS, surpassing the existing methods documented in the literature. The code for this work will be made available at https://github.com/River-ding/LiteMSNet .