Image and Video Completion by Using Bayesian Tensor Decomposition

Reconstruction of image and video from sparse observations attract a great deal of interest in the filed of image/video compression, feature extraction and denoising. Since the color image and video data can be naturally expressed as a tensor structure, many methods based on tensor algebra have been...

Full description

Saved in:
Bibliographic Details
Published inInternational journal of computer science issues Vol. 15; no. 5; pp. 1 - 8
Main Authors Gui, Lihua, Zhao, Xuyang, Zhao, Qibin, Cao, Jianting
Format Journal Article
LanguageEnglish
Published Mahebourg International Journal of Computer Science Issues (IJCSI) 2018
Subjects
Algorithms
Bayesian analysis
Computer science
Decomposition
Feature extraction
Image coding
Image compression
Image reconstruction
Lagrange multiplier
Methods
Noise reduction
Performance prediction
Principal components analysis
Quantitative psychology
Signal processing
Tensors
Video compression
Video data
Online AccessGet full text

Cover

More Information
Summary:Reconstruction of image and video from sparse observations attract a great deal of interest in the filed of image/video compression, feature extraction and denoising. Since the color image and video data can be naturally expressed as a tensor structure, many methods based on tensor algebra have been studied together with promising predictive performance. However, one challenging problem in those methods is tuning parameters empirically which usually requires computational demanding cross validation or intuitive selection. In this paper, we introduce Bayesian Tucker decomposition to reconstruct image and video data from incomplete observation. By specifying the sparsity priors over factor matrices and core tensor, the tensor rank can be automatically inferred via variational bayesian, which greatly reduce the computational cost for model selection. We conduct several experiments on image and video data, which shows that our method outperforms the other tensor methods in terms of completion performance.
ISSN:1694-0814
1694-0784
DOI:10.5281/zenodo.1467644