Multilevel feature fusion dilated convolutional network for semantic segmentation | |
Ku T(库涛)1,2; Yang QR(杨琦瑞)1,2,3; Zhang H(张浩)1,2 | |
刊名 | International Journal of Advanced Robotic Systems |
2021 | |
卷号 | 18期号:2页码:1-12 |
关键词 | Semantic segmentation convolutional neural network deep learning computer vision robot vision |
ISSN号 | 1729-8806 |
产权排序 | 1 |
英文摘要 | Recently, convolutional neural network (CNN) has led to significant improvement in the field of computer vision, especially the improvement of the accuracy and speed of semantic segmentation tasks, which greatly improved robot scene perception. In this article, we propose a multilevel feature fusion dilated convolution network (Refine-DeepLab). By improving the space pyramid pooling structure, we propose a multiscale hybrid dilated convolution module, which captures the rich context information and effectively alleviates the contradiction between the receptive field size and the dilated convolution operation. At the same time, the high-level semantic information and low-level semantic information obtained through multi-level and multi-scale feature extraction can effectively improve the capture of global information and improve the performance of large-scale target segmentation. The encoder–decoder gradually recovers spatial information while capturing high-level semantic information, resulting in sharper object boundaries. Extensive experiments verify the effectiveness of our proposed Refine-DeepLab model, evaluate our approaches thoroughly on the PASCAL VOC 2012 data set without MS COCO data set pretraining, and achieve a state-of-art result of 81.73% mean interaction-over-union in the validate set. |
资助项目 | National Key Research and Development Program of China[2020YFB1708503] |
WOS研究方向 | Robotics |
语种 | 英语 |
WOS记录号 | WOS:000641616600001 |
资助机构 | National Key Research and Development Program of China under grant no. 2020YFB1708503 |
内容类型 | 期刊论文 |
源URL | [http://ir.sia.cn/handle/173321/28766] |
专题 | 沈阳自动化研究所_数字工厂研究室 |
通讯作者 | Yang QR(杨琦瑞) |
作者单位 | 1.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, China 2.Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China 3.University of Chinese Academy of Sciences, Beijing, China |
推荐引用方式 GB/T 7714 | Ku T,Yang QR,Zhang H. Multilevel feature fusion dilated convolutional network for semantic segmentation[J]. International Journal of Advanced Robotic Systems,2021,18(2):1-12. |
APA | Ku T,Yang QR,&Zhang H.(2021).Multilevel feature fusion dilated convolutional network for semantic segmentation.International Journal of Advanced Robotic Systems,18(2),1-12. |
MLA | Ku T,et al."Multilevel feature fusion dilated convolutional network for semantic segmentation".International Journal of Advanced Robotic Systems 18.2(2021):1-12. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论