Refined TV-L-1 Optical Flow Estimation Using Joint Filtering

doi:10.1109/TMM.2019.2929934

CORC > 自动化研究所 > 中国科学院自动化研究所 > 管理与支撑部门 > 科技处

	Refined TV-L-1 Optical Flow Estimation Using Joint Filtering
	Zhang, Congxuan 2,5; Ge, Liyue 5; Chen, Zhen 5; Li, Ming 1; Liu, Wen 3,4; Chen, Hao 1
刊名	IEEE TRANSACTIONS ON MULTIMEDIA
	2020-02-01
卷号	22 期号:2 页码:349-364
关键词	Optical flow TV-L-1 model joint filtering mutual-structure edge-preserving
ISSN号	1520-9210
DOI	10.1109/TMM.2019.2929934
通讯作者	Chen, Zhen(dr_chenzhen@163.com)
英文摘要	Though the accuracy and robustness of optical flow has been dramatically enhanced over the past few years, the issue of edge-blurring near the image and motion boundaries has remained a challenge in flow field estimation. In this paper, we propose a refined total variation with L-1 norm (TV-L-1) optical flow estimation approach using joint filtering, named JOF. First, we divide the image into three categorized regions: mutual-structure regions, inconsistent structure regions, and smooth regions. The mutual-structure guided filter for optical flow estimation is constructed by extracting the mutual-structure regions of the flow field. Second, the refined TV-L-1 optical flow model is proposed by incorporating the non-local term and mutual-structure guided filter objective function into the classical TV-L-1 energy function. Furthermore, the novel TV-L-1 optical flow objective function is minimized using a joint filtering program composed of a weighted median filter and a mutual-structure guided filter to optimize the estimated flow field during the coarse-to-fine optical flow computation scheme. Finally, we compare the proposed JOF method with several state-of-the-art approaches including variational and deep learning based optical flow models using the Middlebury, MPI-Sintel, and UCF101 test databases. The evaluation results indicate that the proposed method has high accuracy and good robustness for flow field computation and, especially, the significant benefit of edge-preserving.
资助项目	National Natural Science Foundation of China[61866026] ; National Natural Science Foundation of China[61772255] ; National Natural Science Foundation of China[61866025] ; Aeronautical Science Foundation of China[2018ZC56008] ; Natural Science Foundation of Jiangxi Province[20171BAB212012] ; Natural Science Foundation of Jiangxi Province[20192BCB23011] ; China Postdoctoral Science Foundation[2019M650894] ; Advantage SubjectTeam Project of Jiangxi Province[20165BCB19007] ; Advantage SubjectTeam Project of Jiangxi Province[20152BCB24004]
WOS关键词	QUALITY ASSESSMENT ; ALGORITHM ; OCCLUSION
WOS研究方向	Computer Science ; Telecommunications
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000510676300006
资助机构	National Natural Science Foundation of China ; Aeronautical Science Foundation of China ; Natural Science Foundation of Jiangxi Province ; China Postdoctoral Science Foundation ; Advantage SubjectTeam Project of Jiangxi Province
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/38486]
专题	自动化研究所_管理与支撑部门_科技处
通讯作者	Chen, Zhen
作者单位	1.Nanchang Hangkong Univ, Key Lab Jiangxi Prov Image Proc & Pattern Recogni, Nanchang 330063, Jiangxi, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 3.Univ Kansas, Bioengn Program, Lawrence, KS 66045 USA 4.Univ Kansas, Dept Phys Therapy & Rehabil Sci, Lawrence, KS 66045 USA 5.Nanchang Hangkong Univ, Key Lab Nondestruct Testing, Minist Educ, Nanchang 330063, Jiangxi, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Congxuan,Ge, Liyue,Chen, Zhen,et al. Refined TV-L-1 Optical Flow Estimation Using Joint Filtering[J]. IEEE TRANSACTIONS ON MULTIMEDIA,2020,22(2):349-364.
APA	Zhang, Congxuan,Ge, Liyue,Chen, Zhen,Li, Ming,Liu, Wen,&Chen, Hao.(2020).Refined TV-L-1 Optical Flow Estimation Using Joint Filtering.IEEE TRANSACTIONS ON MULTIMEDIA,22(2),349-364.
MLA	Zhang, Congxuan,et al."Refined TV-L-1 Optical Flow Estimation Using Joint Filtering".IEEE TRANSACTIONS ON MULTIMEDIA 22.2(2020):349-364.