基于区域自我注意力的实时语义分割网络  

Real-Time Semantic Segmentation Network Based on Regional Self-Attention

作　　者：鲍海龙 万敏[1] 刘忠祥 秦勉 崔浩宇 Bao Hailong;Wan Min;Liu Zhongxiang;Qin Mian;Cui Haoyu(School of Mechatronic Engineering,Southwest Petroleum University,Chengdu,Sichuan 610500,China)

机构地区：[1]西南石油大学机电工程学院,四川成都610500

出　　处：《激光与光电子学进展》2021年第8期196-202,共7页Laser & Optoelectronics Progress

摘　　要：高精度的语义分割结果往往依赖于丰富的空间语义信息与细节信息,但这两者的计算量均较大。为了解决该问题,通过分析图像局部像素具有的相似性,提出了一种基于区域自我注意力的实时语义分割网络。该网络可分别通过一个区域级的自我注意力模块和一个局部交互通道注意力模块计算出特征信息的区域级关联性和通道注意力信息,然后以较少的计算量获取丰富的注意力信息。在Cityscapes数据集上的实验结果表明,相比现有的实时分割网络,本网络的分割精度更高、速度更快。High accuracy results of semantic segmentation often rely on rich spatial semantic information and detailed information,but both incurring high computational costs.In order to solve this problem,we propose a realtime semantic segmentation network based on regional self-attention by observing the similarity of local pixels in the image.The network can calculate the regional correlation of feature information and channel attention information through a regional self-attention module and a local interactive channel attention module.Then,it obtains rich attention information with less calculation.The experimental results on the Cityscapes dataset show that the segmentation accuracy and speed of the network are higher than the existing real-time segmentation network.

关 键 词：图像处理 语义分割 卷积神经网络 注意力机制 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]