Semantic Segmentation Model Based on Adaptive Fusion and Attention Refinement

Authors

Yun Wei, School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, ChinaFollow
Qi Luo, School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, ChinaFollow
Yingzhi Zhao, School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

Abstract

Aiming at the insufficient use of context information and loss of detail information of the existing semantic segmentation, a model based on adaptive fusion and attention refinement is proposed.The model introduces an adaptive fusion module in the process of coding, and solves the insufficient use of context information by fusing each feature map according to the corresponding weight. An attention thinning module is designed in the process of decoding, so that the low-order features and high-order features can guide and optimize each other to solve the loss of detail information.The experimental results show that the average intersection union ratio of the model on PASCAL VOC 2012 dataset reaches 83.7%, which is 1.1% higher than the semantic segmentation model based on encoding and decoding. The average intersection union ratio of 81.7% is obtained on cityscapes dataset, which further verifies the generalization of the model.

Recommended Citation

Wei, Yun; Luo, Qi; and Zhao, Yingzhi (2023) "Semantic Segmentation Model Based on Adaptive Fusion and Attention Refinement," Journal of System Simulation: Vol. 35: Iss. 6, Article 9.
DOI: 10.16182/j.issn1004731x.joss.22-0169
Available at: https://dc-china-simulation.researchcommons.org/journal/vol35/iss6/9

First Page

1226

Revised Date

2022-03-21

DOI Link

http://dx.doi.org/10.16182/j.issn1004731x.joss.22-0169

Last Page

1234

CLC

TP391

Recommended Citation

Yun Wei, Qi Luo, Yingzhi Zhao. Semantic Segmentation Model Based on Adaptive Fusion and Attention Refinement[J]. Journal of System Simulation, 2023, 35(6): 1226-1234.

DOI

10.16182/j.issn1004731x.joss.22-0169