Field-of-Experts Filters Guided Tensor Completion

• Published in: IEEE transactions on multimedia Vol. 20; no. 9; pp. 2316 - 2329
• Main Authors: Xiong, Biao, Liu, Qiegen, Xiong, Jiaojiao, Li, Sanqian, Wang, Shanshan, Liang, Dong
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.09.2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: alternative minimization; Brain modeling; Data models; field-of-experts; Low-rank tensor; Matrix decomposition; Minimization; multi-view features; Target recognition; Tensile stress; tensor completion; Visualization
• ISSN: 1520-9210; 1941-0077
• DOI: 10.1109/TMM.2018.2806225

Most low-rank tensor approximations are NP-hard problems. In this paper, we introduce a novel concept: field-of-experts (FoE) filters guided tensor completion, which aims to integrate the strengths of the emerging tensor completion method and the conventional FoE filters. Specifically, the target image is convolved by FoE filters to produce multiview features as a high-order tensor, which captures complementary information from multiple views. In order to impose the concept, we employ two strategies to model the new tensor, one is called FoE filters guided low-rank tensor completion, and another is called FoE filters guided simultaneous tensor decomposition and completion (FoE-STDC). The resulting objectives are solved efficiently by alternating minimization. Extensive experimental results validate the superior performance and robustness of the proposed methods over their corresponding counterparts in all cases. Particularly, the proposed FoE-STDC is superior to the state-of-the-art tensor completion methods.