RTSNet: Real-Time Semantic Segmentation Network for Outdoor Scenes

	RTSNet: Real-Time Semantic Segmentation Network for Outdoor Scenes
	Ma, Mingyu 2; Zou FS(邹风山)2,3; Xu F(徐方)1,2,3; Song JL(宋吉来)1,3
	2019
会议日期	July 29 - August 2, 2019
会议地点	Suzhou, China
关键词	semantic segmentation real-time outdoor scenes RTSNet mean intersection-over-union
页码	659-664
英文摘要	Semantic segmentation technique plays an important role in robotics related applications, especially autonomous driving and assisted driving. Real-time semantic segmentation has very significant practical meaning, but many studies focus on accuracy, not computationally efficient solutions. In this paper, a real-time semantic segmentation network based on encoder-decoder architecture is proposed. This framework's encoder part adopted a lightweight network architecture for feature extraction and this architecture is mainly based on the MobilenetV2. Its decoder part is decided to use the Skip architecture. This architecture can utilize higher resolution feature mapping to provide adequate accuracy and greatly improve computational efficiency. We evaluated RTSNet on the Cityscapes dataset for urban scenes and compared with the state of the art real-time semantic segmentation networks. The mean intersection-over-union it can achieve on the Cityscapes dataset is about 62.0%, while it achieved 14.0 fps on NVIDIA Jetson TX2 with 360×640 input images.
产权排序	3
会议录	Proceedings of 9th IEEE International Conference on CYBER Technology in Automation, Control, and Intelligent Systems
会议录出版者	IEEE
会议录出版地	New York
语种	英语
ISBN号	978-1-7281-0769-1
WOS记录号	WOS:000569550300113
内容类型	会议论文
源URL	[http://ir.sia.cn/handle/173321/26836]
专题	沈阳自动化研究所_其他
通讯作者	Ma, Mingyu
作者单位	1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 10016, China 2.Northeastern University, Shenyang 110819, China 3.Shenyang SIASUN Robot Automation Co. Ltd., Shenyang 110168, China
推荐引用方式 GB/T 7714	Ma, Mingyu,Zou FS,Xu F,et al. RTSNet: Real-Time Semantic Segmentation Network for Outdoor Scenes[C]. 见:. Suzhou, China. July 29 - August 2, 2019.