arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1903.04735cs.LGstat.ML
收藏

低秩张量网格的图像补齐算法

Low-rank Tensor Grid for Image Completion

University of Electronic Science and Technology of China(电子科技大学)

Huyan Huang, Yipeng Liu, Ce Zhu

摘要:

张量补全通过利用多路数据的低秩结构来估计缺失的分量。最近提出的基于张量训练(TT)和张量环(TR)的方法在图像恢复方面表现出比经典方法更好的性能。与TT和TR相比,投影纠缠对态(PEPS),也称为张量网格(TG),允许更多的不同维度之间的相互作用,并可能导致更紧凑的表示。在本文中,我们提出了低秩张量网格的基础上进行图像完成。采用两阶段密度矩阵重整化群算法对由多个TT分解组成的TG分解进行初始化。通过求解交替最小二乘问题,可以交替地获得潜在TG因子。为了进一步提高计算效率,采用并行矩阵分解的方法,提出了一种低秩TG完备化的多线性矩阵分解方法。在合成数据和真实图像上的实验结果表明,所提出的方法在恢复精度方面优于现有的方法。

英文摘要:

Tensor completion estimates missing components by exploiting the low-rank structure of multi-way data. The recently proposed methods based on tensor train (TT) and tensor ring (TR) show better performance in image recovery than classical ones. Compared with TT and TR, the projected entangled pair state (PEPS), which is also called tensor grid (TG), allows more interactions between different dimensions, and may lead to more compact representation. In this paper, we propose to perform image completion based on low-rank tensor grid. A two-stage density matrix renormalization group algorithm is used for initialization of TG decomposition, which consists of multiple TT decompositions. The latent TG factors can be alternatively obtained by solving alternating least squares problems. To further improve the computational efficiency, a multi-linear matrix factorization for low rank TG completion is developed by using parallel matrix factorization. Experimental results on synthetic data and real-world images show the proposed methods outperform the existing ones in terms of recovery accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑