Exploiting Lightweight Statistical Learning for Event-Based Vision Processing

doi:10.1109/ACCESS.2018.2823260

	Exploiting Lightweight Statistical Learning for Event-Based Vision Processing
	Shi, Cong 1; Li, Jiajun 2; Wang, Ying 2; Luo, Gang 1
	2018
发表期刊	IEEE ACCESS
ISSN	2169-3536
卷号	6 页码:19396-19406
摘要	This paper presents a lightweight statistical learning framework potentially suitable for low-cost event-based vision systems, where visual information is captured by a dynamic vision sensor (DVS) and represented as an asynchronous stream of pixel addresses (events) indicating a relative intensity change on those locations. A simple random ferns classifier based on randomly selected patch-based binary features is employed to categorize pixel event flows. Our experimental results demonstrate that compared to existing event-based processing algorithms, such as spiking convolutional neural networks (SCNNs) and the state-of-the-art bag-of-events (BoE)-based statistical algorithms, our framework excels in high processing speed (2x faster than the BoE statistical methods and >100x faster than previous SCNNs in training speed) with extremely simple online learning process, and achieves state-of-the-art classification accuracy on four popular address-event representation data sets: MNIST-DVS, Poker-DVS, Posture-DVS, and CIFAR10-DVS. Hardware estimation shows that our algorithm will be preferable for low-cost embedded system implementations.
关键词	Address-event representation (AER) dynamic vision sensor (DVS) random ferns statistical learning neuromorphic processing
DOI	10.1109/ACCESS.2018.2823260
收录类别	SCI
语种	英语
资助项目	NIH[R01 AG041794]
WOS研究方向	Computer Science ; Engineering ; Telecommunications
WOS类目	Computer Science, Information Systems ; Engineering, Electrical & Electronic ; Telecommunications
WOS记录号	WOS:000430998000001
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：8[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/2XEOYT63/5351
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Wang, Ying
作者单位	1.Harvard Med Sch, Schepens Eye Res Inst, Massachusetts Eye & Ear, Dept Ophthalmol, Boston, MA 02114 USA 2.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100864, Peoples R China
推荐引用方式 GB/T 7714	Shi, Cong,Li, Jiajun,Wang, Ying,et al. Exploiting Lightweight Statistical Learning for Event-Based Vision Processing[J]. IEEE ACCESS,2018,6:19396-19406.
APA	Shi, Cong,Li, Jiajun,Wang, Ying,&Luo, Gang.(2018).Exploiting Lightweight Statistical Learning for Event-Based Vision Processing.IEEE ACCESS,6,19396-19406.
MLA	Shi, Cong,et al."Exploiting Lightweight Statistical Learning for Event-Based Vision Processing".IEEE ACCESS 6(2018):19396-19406.