Visual Data Completion via Local Sensitive Low Rank Tensor Learning

被引:0
作者
Liu, Qing-Yi [1 ]
Zhu, Lin [1 ]
Huang, De-Shuang [1 ]
Wang, Gang [1 ]
Zhang, Xiao-Ping [1 ]
Shang, Li [2 ]
Huang, Zhi-Kai [3 ]
机构
[1] Tongji Univ, Coll Elect & Informat Engn, Inst Machine Learning & Syst Biol, Caoan Rd 4800, Shanghai 201804, Peoples R China
[2] Suzhou Vocat Univ, Coll Elect Informat Engn, Dept Commun Technol, Suzhou 215104, Jiangsu, Peoples R China
[3] Nanchang Inst Technol, Coll Mech & Elect Engn, Nanchang 330099, Jiangxi, Peoples R China
来源
2016 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN) | 2016年
基金
中国博士后科学基金; 美国国家科学基金会;
关键词
Tensor; Data completion; Low rank; Trace norm; Tensor completion;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The problems of estimating missing values in visual data appear ubiquitously in computer vision applications including image inpainting, video inpainting, hyperspectral data recovery, and magnetic resonance imaging (MRI) data recovery. Recently, it is shown that tensor completion, which generalizes matrix completion to multiway data of higher order, could accurately estimate the overall data structure and achieves state-of-the-art performance for video completion. However, current tensor-based approaches implicitly assume that the partially observed video is globally low rank, which is too stringent for practical applications where the input video could include multiple heterogeneous episodes and the global correlation between frames is not high. To tackle this problem, we propose a novel local sensitive formulation of tensor learning where we assume instead that the video is inter-correlated in a local manner, leading to a representation of the observed tensor as a weighted sum of low-rank tensors. Computationally, we also design efficient scheme for solving the resulting learning problem based on the alternating direction method of multipliers (ADMM). Our experiments show improvements in prediction accuracy over classical approaches for visual data completion tasks.
引用
收藏
页码:2693 / 2698
页数:6
相关论文
共 22 条
[1]  
[Anonymous], 2006, 2006 IEEE COMP SOC C
[2]  
[Anonymous], 2007, P KDD CUP WORKSH
[3]  
[Anonymous], 2006, KDD
[4]  
[Anonymous], 2005, 5 IEEE INT C DAT MIN
[5]  
[Anonymous], 2008, ACM Transactions on Knowledge Discovery from Data (TKDD), DOI DOI 10.1145/1409620.1409621
[6]   Image inpainting [J].
Bertalmio, M ;
Sapiro, G ;
Caselles, V ;
Ballester, C .
SIGGRAPH 2000 CONFERENCE PROCEEDINGS, 2000, :417-424
[7]   A SINGULAR VALUE THRESHOLDING ALGORITHM FOR MATRIX COMPLETION [J].
Cai, Jian-Feng ;
Candes, Emmanuel J. ;
Shen, Zuowei .
SIAM JOURNAL ON OPTIMIZATION, 2010, 20 (04) :1956-1982
[8]  
FAZEL M, 2001, ACC, P4734
[9]  
Gabay D., 1976, Computers & Mathematics with Applications, V2, P17, DOI 10.1016/0898-1221(76)90003-1
[10]  
Gandy S., 2011, INVERSE PROBL, P27