Fast Tensor Nuclear Norm for Structured Low-Rank Visual Inpainting

Xu, Honghui; Zheng, Jianwei; Yao, Xiaomin; Feng, Yuchao; Chen, Shengyong

doi:10.1109/tcsvt.2021.3067022

Cited by 35 publications

(7 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that the truncated tubal nuclear norm [40] can also be used in the formulation (20). Similar tensor completion formulation is used in [41] but here we have used unitary transform matrices instead of discrete Fourier transform matrix that is used in the traditional tensor SVD and has shown to provide better results [42]. We also proposed to improve the image recovery by smoothing the results at each iteration (Line 5 in Algorithm 2).…”

Section: B Tensor Completion Using Tensor Nuclear Norm (Tnn) Regulari...mentioning

confidence: 99%

Matrix and tensor completion using tensor ring decomposition with sparse representation

Asante-Mensah

Ahmadi‐Asl

Cichocki

2021

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

Completing a data tensor with structured missing components is a challenging task where the missing components are not distributed randomly but they admit some regular patterns, e.g. missing columns and rows or missing blocks/patches. Many of the existing tensor completion algorithms are not able to handle such scenarios. In this paper, we propose a novel and efficient approach for matrix/tensor completion by applying Hankelization and distributed tensor ring decomposition. Our main idea is first Hankelizing an incomplete data tensor in order to obtain high-order tensors and then completing the data tensor by imposing sparse representation on the core tensors in tensor ring format. We apply an efficient over-complete discrete cosine transform dictionary and sparse representation techniques to learn core tensors. Alternating direction methods of multiplier and accelerated proximal gradient approaches are used to solve the underlying optimization problems. Extensive simulations performed on image, video completions and time series forecasting show the validity and applicability of the method for different kinds of structured and random missing elements.

show abstract

Section: B Tensor Completion Using Tensor Nuclear Norm (Tnn) Regulari...mentioning

confidence: 99%

Matrix and tensor completion using tensor ring decomposition with sparse representation

Asante-Mensah

Ahmadi‐Asl

Cichocki

2021

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…In the restoration of color images and videos, the tensor tubal rank model based on the tensor-tensor product and tensor singular value decomposition (t-SVD) shows better performance than other rank models. The definitions of the tensor-tensor product, tensor singular value decomposition (t-SVD), tensor tubal rank, and tensor nuclear norm can be found in [41,42].…”

Section: Snow Video Background Modelingmentioning

confidence: 99%

Video Desnowing and Deraining via Saliency and Dual Adaptive Spatiotemporal Filtering

Jia

et al. 2021

Sensors

View full text Add to dashboard Cite

Outdoor vision sensing systems often struggle with poor weather conditions, such as snow and rain, which poses a great challenge to existing video desnowing and deraining methods. In this paper, we propose a novel video desnowing and deraining model that utilizes the salience information of moving objects to address this problem. First, we remove the snow and rain from the video by low-rank tensor decomposition, which makes full use of the spatial location information and the correlation between the three channels of the color video. Second, because existing algorithms often regard sparse snowflakes and rain streaks as moving objects, this paper injects salience information into moving object detection, which reduces the false alarms and missed alarms of moving objects. At the same time, feature point matching is used to mine the redundant information of moving objects in continuous frames, and a dual adaptive minimum filtering algorithm in the spatiotemporal domain is proposed by us to remove snow and rain in front of moving objects. Both qualitative and quantitative experimental results show that the proposed algorithm is more competitive than other state-of-the-art snow and rain removal methods.

show abstract

“…Albeit attaining encouraging performance, the DL-based super-resolution methods still suffer from some drawbacks. The most critical one is the ignorance of the explicit priors, which are known as the inherent properties of most HSIs, e.g., spectral low-rank property [40]. This may lead to a deviation of the network output from the general prior configuration and impact negatively on the recovery accuracy.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Image Superresolution via Subspace-Based Deep Prior Regularization

Zheng

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Hyperspectral imaging is able to provide a finer delivery of various material properties than conventional imaging systems. Yet in reality, an optical system can only generate data with high spatial resolution but low spectral one, or vice versa, at video rates. As a result, an issue that fuses lowresolution hyperspectral (LRHS) and high-resolution multispectral (HRMS) images has gained great attention. However, most fusion approaches depend purely on hand-crafted regularizers or data-driven priors, leading to the issues of tricky parameter selection or poor interpretability. In this work, a subspace-based deep prior regularization (SDPR) is proposed to tackle these problems, which takes both hand-crafted regularizer and datadriven prior into account. Specifically, we leverage the spectral correlation of the images and transfer them from the original space to the subspace domain, within which a modified U-netbased deep prior learning network (SDPL-net) is designed for the fusion issue. Moreover, instead of taking the output of SDPL-net directly as the result, we further feed the output back to the model-based optimization. Under such prior regularization, the recovered high-resolution hyperspectral (HRHS) image holds a high consistency to its inherent structure and hence tends to present enhanced reliability and accuracy. Experimental results on simulated and real data reveal that the proposed method excels other state-of-the-art (SOTA) methods in both quantitative and qualitative metrics.

show abstract

Fast Tensor Nuclear Norm for Structured Low-Rank Visual Inpainting

Cited by 35 publications

References 44 publications

Matrix and tensor completion using tensor ring decomposition with sparse representation

Matrix and tensor completion using tensor ring decomposition with sparse representation

Video Desnowing and Deraining via Saliency and Dual Adaptive Spatiotemporal Filtering

Hyperspectral Image Superresolution via Subspace-Based Deep Prior Regularization

Contact Info

Product

Resources

About