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. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Accelerating DNN-based 3D point cloud processing for mobile computing

文献类型：期刊论文

作者Liu, Bosheng1,2; Chen, Xiaoming1; Han, Yinhe1; Li, Jiajun1,2; Xu, Haobo1,2; Li, Xiaowei1
刊名SCIENCE CHINA-INFORMATION SCIENCES
出版日期2019-11-01
卷号62期号:11页码:11
ISSN号1674-733X
关键词deep neural network acceleration point cloud data neighbor point search mobile robotics hardware architecture
DOI10.1007/s11432-019-9932-3
英文摘要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.
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
语种英语
出版者SCIENCE PRESS
WOS记录号WOS:000487971900001
源URL[http://119.78.100.204/handle/2XEOYT63/4588]  
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.

入库方式： OAI收割

来源：计算技术研究所

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。