Context-adaptive matching for optical flow

doi:10.1007/s11042-017-5386-2

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	Context-adaptive matching for optical flow
	Bao, Xiuguo 3; Zu, Yueran 4; Gao, Ke 1; Wang, Yanyang 2; Tang, Wenzhong 4
刊名	MULTIMEDIA TOOLS AND APPLICATIONS
	2019
卷号	78 期号:1 页码:641-659
关键词	Optical flow PatchMatch Edge preserving Large displacement
ISSN号	1380-7501
DOI	10.1007/s11042-017-5386-2
英文摘要	Modern sparse-to-dense optical flow estimation algorithms usually achieve state-of-art performance. Those algorithms need two steps: matching and interpolation. Matching is often unreliable for very large displacement optical flow due to illumination changes, deformations and occlusion etc. Moreover, conspicuous errors around motion discontinuities still keep serious as most methods consider edge only at interpolation step. The context-adaptive matching (CAM) is proposed for optical flow which is better at large displacement and edge preserving. The CAM is selective in feature extraction, adaptive in flow propagation and search radius adjusting. Selective features are proposed to consider edge preserving in matching step. Except for the usually used SIFT descriptor, the local directional pattern flow (LDPF) is introduced to keep more edge structure, and the oriented fast and rotated brief (ORB) is utilized to select out several most similar candidates. Unlike coarse-to-fine matching, which proposed a propagation step with only neighbors, we propose adaptive propagation to extend the matching candidates in order to improve the possibility of getting right correspondences. Furthermore, guided by prior knowledge and taking advantage of upper layers results, adaptive radius instead of constrained radius are proposed at finer layers. The CAM interpolated by EpicFlow is fast and robust for large displacements especially for fast moving objects and also preserves the edge structure well. Extensive experiments show that our algorithm is on par with the state-of-art optical flow methods on MPI-Sintel, KITTI and Middlebury.
资助项目	Beijing Municipal Science and Technology Commission Project[Z171100000117010] ; National Key Research and Development Plan[2016YFB0801203] ; National Key Research and Development Plan[2016YFB0801200] ; National Nature Science Foundation of China[61271428]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	SPRINGER
WOS记录号	WOS:000457317500036
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/3458]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Gao, Ke
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing 100080, Peoples R China 2.Beihang Univ, Sch Aeronaut Sci & Engn, Beijing 100191, Peoples R China 3.Coordinat Ctr China CNCERT, Natl Comp Network Emergency Response Tech Team, Beijing 100029, Peoples R China 4.Beihang Univ, Sch Comp Sci & Engn, Beijing 100191, Peoples R China
推荐引用方式 GB/T 7714	Bao, Xiuguo,Zu, Yueran,Gao, Ke,et al. Context-adaptive matching for optical flow[J]. MULTIMEDIA TOOLS AND APPLICATIONS,2019,78(1):641-659.
APA	Bao, Xiuguo,Zu, Yueran,Gao, Ke,Wang, Yanyang,&Tang, Wenzhong.(2019).Context-adaptive matching for optical flow.MULTIMEDIA TOOLS AND APPLICATIONS,78(1),641-659.
MLA	Bao, Xiuguo,et al."Context-adaptive matching for optical flow".MULTIMEDIA TOOLS AND APPLICATIONS 78.1(2019):641-659.