Dense context distillation network for semantic parsing of oblique UAV images

doi:10.1016/j.jag.2022.103062

	Dense context distillation network for semantic parsing of oblique UAV images
	Ding, Youli 1; Zheng, Xianwei 1; Chen, Yiping 2; Shen, Shuhan3 ; Xiong, Hanjiang 1
发表期刊	INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
ISSN	1569-8432
	2022-11-01
卷号	114 页码:12
通讯作者	Zheng, Xianwei(zhengxw@whu.edu.cn)
摘要	Semantic segmentation of oblique unmanned aerial vehicle (UAV) images serves as a foundation for many modern urban applications, such as road scene monitoring and semantic 3D modeling. However, objects in UAV images can vary intensely in size and undergo severe perspective distortion because of the oblique viewing style. Existing general segmentation models designed for ground and remote sensing images rarely considered these challenges specific to UAV images. Therefore, they have large difficulties in learning discriminative representation for simultaneously reasoning the extremely large and small objects in UAV images. In this paper, we propose a dense context distillation network (DCDNet) to learn distortion-robust feature representation for semantic segmentation of UAV images. The basic DCDNet is deployed as an dual-branch encoder-decoder architecture. To accomplish the goal of dense context distillation, DCDNet is first equipped with several cross -scale context selectors at different encoding stages to densely and selectively gather the useful context from low -to high-level dual-scale feature maps. A joint supervision is then applied to reinforce the learning of shallower features for distilling more low-level contexts that are vital to the reasoning of small or thin structures. A multi-scale feature aggregator is incorporated to adaptively fuse the long-range context during decoding, which absorbs the complementary merits of the dense context captured from feature maps of different levels. With the dense context distillation, DCDNet is more capable of offering the differently scaled objects with the required context for better learning and prediction. Extensive experiments on the challenging UAVid dataset demonstrate that our DCDNet can well adapt to the oblique UAV images, achieving a state-of-the-art segmentation performance with a mIoU score of 72.38%.
关键词	UAV Road scene Semantic segmentation Deep learning Dense context
DOI	10.1016/j.jag.2022.103062
关键词[WOS]	SEGMENTATION ; AGGREGATION ; RGB
收录类别	SCI
语种	英语
资助项目	Open fund of Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources[KF202106084] ; National Natural Science Foundation of China[41871361] ; National Natural Science Foundation of China[42071370] ; Fundamental Research Funds for the Central Universities[2042022kf1203]
项目资助者	Open fund of Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources ; National Natural Science Foundation of China ; Fundamental Research Funds for the Central Universities
WOS研究方向	Remote Sensing
WOS类目	Remote Sensing
WOS记录号	WOS:000907048500002
出版者	ELSEVIER
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/51169
专题	多模态人工智能系统全国重点实验室_机器人视觉
通讯作者	Zheng, Xianwei
作者单位	1.Wuhan Univ, State Key Lab LIESMARS, Wuhan 430079, Peoples R China 2.Sun Yat Sen Univ, Sch Geospatial Engn & Sci, Zhuhai 519082, Peoples R China 3.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Ding, Youli,Zheng, Xianwei,Chen, Yiping,et al. Dense context distillation network for semantic parsing of oblique UAV images[J]. INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION,2022,114:12.
APA	Ding, Youli,Zheng, Xianwei,Chen, Yiping,Shen, Shuhan,&Xiong, Hanjiang.(2022).Dense context distillation network for semantic parsing of oblique UAV images.INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION,114,12.
MLA	Ding, Youli,et al."Dense context distillation network for semantic parsing of oblique UAV images".INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION 114(2022):12.