空间通道双重注意力道路场景语义分割  

Semantic Segmentation of Unmanned Driving Scene Based on Spatial Channel Dual Attention

作　　者：王小玉[1] 林鹏 WANG Xiaoyu;LIN Peng(Harbin University of Scienceand Technology,Computer Scienceand Technology,Harbin 150080,China)

机构地区：[1]哈尔滨理工大学计算机科学与技术学院,哈尔滨150080

出　　处：《哈尔滨理工大学学报》2023年第5期103-109,共7页Journal of Harbin University of Science and Technology

基　　金：国家自然科学基金(61772160);黑龙江省教育厅科学技术研究项目(12541177)。

摘　　要：无人驾驶领域的一个重要问题就是在低功耗移动电子设备上怎样运行实时高精度语义分割模型。由于现有语义分割算法参数量过多、内存占用巨大导致很难满足无人驾驶等现实应用的问题,并且在影响语义分割模型的精度和推理速度的众多因素中,空间信息和上下文特征尤为重要,并且很难同时兼顾。针对该问题提出采用不完整的ResNet18作为骨干网络,ResNet18是一个轻量级的模型,参数量较少,占用内存不大;同时采用双边语义分割模型的技术,在两条路径上添加通道空间双重注意力机制,来获取更多的上下文信息和空间信息的想法。另外还采用了精炼上下文信息的注意力优化模块,和融合两条路径输出的融合模块,添加的模块对于参数量和内存的影响很小,可以即插即用。以Cityscapes和CamVid为数据集。在Citycapes上,mIoU达到77.3%;在CamVid上,mIoU达到66.5%。输入图像分辨率为1024×2048时,推理时间为37.9 ms。An important issue in the field of unmanned driving is how to run real-time high-precision semantic segmentation models on low-power mobile electronic devices.Existing semantic segmentation algorithms have too many parameters and huge memory usage,which makes it difficult to meet the problems of real-world applications such as unmanned driving.However,among the many factors that affect the accuracy and speed of the semantic segmentation model,spatial information and contextual features are particularly important,and it is difficult to take into account both.In response to this problem,it is proposed to use the incomplete ResNet18 as the backbone network,design a bilateral semantic segmentation model,and add a channel space dual attention model to the two paths to obtain more contextual and spatial information.In addition,the attention optimization module that refines the context information and the fusion module that integrates the output of the two paths are also used.Take Cityscapes and CamVid as data sets.On Citycapes,mIoU reached 77.3%;on CamVid,mIoU reached 66.5%.When the input image resolution is 1024×2048,the segmentation speed is 37.9 ms.

关 键 词：无人驾驶 实时语义分割 深度学习 注意力机制 深度可分离卷积 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]