Accelerating DNN-based 3D point cloud processing for mobile computing

doi:10.1007/s11432-019-9932-3

	Accelerating DNN-based 3D point cloud processing for mobile computing
	Liu, Bosheng 1,2; Chen, Xiaoming 1; Han, Yinhe 1; Li, Jiajun 1,2; Xu, Haobo 1,2; Li, Xiaowei 1
	2019-11-01
发表期刊	SCIENCE CHINA-INFORMATION SCIENCES
ISSN	1674-733X
卷号	62 期号:11 页码:11
摘要	3D point cloud data, which are produced by various 3D sensors such as LIDAR and stereo cameras, have been widely deployed by industry leaders such as Google, Uber, Tesla, and Mobileye, for mobile robotic applications such as autonomous driving and humanoid robots. Point cloud data, which are composed of reliable depth information, can provide accurate location and shape characteristics for scene understanding, such as object recognition and semantic segmentation. However, deep neural networks (DNNs), which directly consume point cloud data, are particularly computation-intensive because they have to not only perform multiplication-and-accumulation (MAC) operations but also search neighbors from the irregular 3D point cloud data. Such a task goes beyond the capabilities of general-purpose processors in realtime to figure out the solution as the scales of both point cloud data and DNNs increase from application to application. We present the first accelerator architecture that dynamically configures the hardware on-the-fly to match the computation of both neighbor point search and MAC computation for point-based DNNs. To facilitate the process of neighbor point search and reduce the computation costs, a grid-based algorithm is introduced to search neighbor points from a local region of grids. Evaluation results based on the scene recognition and segmentation tasks show that the proposed design harvests 16.4 x higher performance and saves 99.95% of energy than an NVIDIA Tesla K40 GPU baseline in point cloud scene understanding applications.
关键词	deep neural network acceleration point cloud data neighbor point search mobile robotics hardware architecture
DOI	10.1007/s11432-019-9932-3
收录类别	SCI
语种	英语
资助项目	China Scholarship Council (CSC) ; National Natural Science Foundation of China[61804155] ; National Natural Science Foundation of China[61522406] ; National Natural Science Foundation of China[61834006] ; National Natural Science Foundation of China[61532017] ; National Natural Science Foundation of China[61521092] ; Beijing Municipal Science & Technology Commission[Z171100000117019] ; Beijing Municipal Science & Technology Commission[Z181100008918006] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDPB12] ; Youth Innovation Promotion Association CAS ; Young Elite Scientists Sponsorship Program by CAST[2018QNRC001] ; Innovative Project of ICT, CAS[5120186140]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Information Systems ; Engineering, Electrical & Electronic
WOS记录号	WOS:000487971900001
出版者	SCIENCE PRESS
引用统计	被引频次：9[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/4589
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Chen, Xiaoming; Han, Yinhe
作者单位	1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Liu, Bosheng,Chen, Xiaoming,Han, Yinhe,et al. Accelerating DNN-based 3D point cloud processing for mobile computing[J]. SCIENCE CHINA-INFORMATION SCIENCES,2019,62(11):11.
APA	Liu, Bosheng,Chen, Xiaoming,Han, Yinhe,Li, Jiajun,Xu, Haobo,&Li, Xiaowei.(2019).Accelerating DNN-based 3D point cloud processing for mobile computing.SCIENCE CHINA-INFORMATION SCIENCES,62(11),11.
MLA	Liu, Bosheng,et al."Accelerating DNN-based 3D point cloud processing for mobile computing".SCIENCE CHINA-INFORMATION SCIENCES 62.11(2019):11.