多层级多尺度特征融合的轻量级语义分割算法  

Lightweight Image Semantic Segmentation Algorithm Based on Multi-Level and Multi-Scale Feature Fusion

作　　者：肖漫漫 赵三元[2] 李浩 XIAO Man-man;ZHAO San-yuan;LI Hao(Information Technology Center,Renmin University of China,Beijing 100872,China;School of Computer Science,Beijing Institute of Technology,Beijing,100081,China)

机构地区：[1]中国人民大学信息技术中心,北京100872 [2]北京理工大学计算机学院,北京100081

出　　处：《计算机仿真》2024年第10期168-173,共6页Computer Simulation

基　　金：国家自然科学基金资助项目(61902027)。

摘　　要：针对自动驾驶视觉图像输入,提出了基于多层级多尺度特征融合的轻量级语义分割算法。算法采用编码器解码器架构,首先研究了低下采样率的编码器骨干网络。网络基于轻量级卷积模块,并采用不同深度的卷积层输出相同分辨率特征图,从而生成多层级的语义特征;其次研究了基于多特征融合的语义分割解码器,解码器设计包含多尺度特征融合模块和多层级特征融合模块两部分,结合多尺度特征和通道注意力机制,提取并融合编码器输出特征,最终生成分割结果。A lightweight semantic segmentation algorithm based on multi-level and multi-scale feature fusion is proposed for autonomous driving visual image input.The algorithm adopts an encoder decoder architecture and first studies the encoder backbone network with low sampling rate.A low sampling rate encoder backbone is proposed based on the lightweight convolution operation.The encoder outputs multi-level feature maps with the same resolution at different depths of convolutional layers.A decoder us proposed based on the multi-feature fusion method,which contains the multi-level feature fusion module and the multi-scale feature fusion module.The decoder aggregates output features of the encoder and generates the segmentation mask.

关 键 词：计算机视觉 图像分割 轻量级骨干网络 多层级特征融合 多尺度特征融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]