English

A feature refinement module for light-weight semantic segmentation network

Computer Vision and Pattern Recognition 2024-12-13 v1 Machine Learning Image and Video Processing

Abstract

Low computational complexity and high segmentation accuracy are both essential to the real-world semantic segmentation tasks. However, to speed up the model inference, most existing approaches tend to design light-weight networks with a very limited number of parameters, leading to a considerable degradation in accuracy due to the decrease of the representation ability of the networks. To solve the problem, this paper proposes a novel semantic segmentation method to improve the capacity of obtaining semantic information for the light-weight network. Specifically, a feature refinement module (FRM) is proposed to extract semantics from multi-stage feature maps generated by the backbone and capture non-local contextual information by utilizing a transformer block. On Cityscapes and Bdd100K datasets, the experimental results demonstrate that the proposed method achieves a promising trade-off between accuracy and computational cost, especially for Cityscapes test set where 80.4% mIoU is achieved and only 214.82 GFLOPs are required.

Keywords

Cite

@article{arxiv.2412.08670,
  title  = {A feature refinement module for light-weight semantic segmentation network},
  author = {Zhiyan Wang and Xin Guo and Song Wang and Peixiao Zheng and Lin Qi},
  journal= {arXiv preprint arXiv:2412.08670},
  year   = {2024}
}

Comments

Accept by icip 2023

R2 v1 2026-06-28T20:31:28.868Z