中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 多模态人工智能系统全国重点实验室
  5. 机器人视觉团队
SCE-Net: Self- and Cross-Enhancement Network for Single-View Height Estimation and Semantic Segmentation

文献类型：期刊论文

作者Xing, Siyuan2,3; Dong, Qiulei1,2,3; Hu, Zhanyi2,3
刊名REMOTE SENSING
出版日期2022-05-01
卷号14期号:9页码:22
关键词height estimation semantic segmentation single aerial image convolutional neural networks multi-task learning deep metric learning
DOI10.3390/rs14092252
通讯作者Dong, Qiulei(qldong@nlpr.ia.ac.cn)
英文摘要Single-view height estimation and semantic segmentation have received increasing attention in recent years and play an important role in the photogrammetry and remote sensing communities. The height information and semantic information of images are correlated, and some recent works have shown that multi-task learning methods can achieve complementation of task-related features and improve the prediction results of the multiple tasks. Although much progress has been made in recent works, how to effectively extract and fuse height features and semantic features is still an open issue. In this paper, a self- and cross-enhancement network (SCE-Net) is proposed to jointly perform height estimation and semantic segmentation on single aerial images. A feature separation-fusion module is constructed to effectively separate and fuse height features and semantic features based on an attention mechanism for feature representation enhancement across tasks. In addition, a height-guided feature distance loss and a semantic-guided feature distance loss are designed based on deep metric learning to achieve task-aware feature representation enhancement. Extensive experiments are conducted on the Vaihingen dataset and the Potsdam dataset to verify the effectiveness of the proposed method. The experimental results demonstrate that the proposed SCE-Net could outperform the state-of-the-art methods and achieve better performance in both height estimation and semantic segmentation.
WOS关键词REMOTE-SENSING IMAGES ; OBJECT DETECTION ; AERIAL IMAGES ; DEEP ; RGB ; CLASSIFICATION ; RECONSTRUCTION ; SURFACE
资助项目National Natural Science Foundation of China[61991423] ; National Natural Science Foundation of China[U1805264] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDB32050100]
WOS研究方向Environmental Sciences & Ecology ; Geology ; Remote Sensing ; Imaging Science & Photographic Technology
语种英语
出版者MDPI
WOS记录号WOS:000794398800001
资助机构National Natural Science Foundation of China ; Strategic Priority Research Program of the Chinese Academy of Sciences
源URL[http://ir.ia.ac.cn/handle/173211/49389]  
专题自动化研究所_模式识别国家重点实验室_机器人视觉团队
通讯作者Dong, Qiulei
作者单位1.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing 100190, Peoples R China
2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
推荐引用方式
GB/T 7714		 Xing, Siyuan,Dong, Qiulei,Hu, Zhanyi. SCE-Net: Self- and Cross-Enhancement Network for Single-View Height Estimation and Semantic Segmentation[J]. REMOTE SENSING,2022,14(9):22.
APA Xing, Siyuan,Dong, Qiulei,&Hu, Zhanyi.(2022).SCE-Net: Self- and Cross-Enhancement Network for Single-View Height Estimation and Semantic Segmentation.REMOTE SENSING,14(9),22.
MLA Xing, Siyuan,et al."SCE-Net: Self- and Cross-Enhancement Network for Single-View Height Estimation and Semantic Segmentation".REMOTE SENSING 14.9(2022):22.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

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。