An Overcomplete Dictionary Learning Algorithm for Tensor Sparse Representation
スポンサーリンク
概要
- 論文の詳細を見る
Recently sparse coding has received expressions of interest in the field of pattern recognition. Most existing methods take the data-as-vector formulation, and deal with images (the second order tensor) or volume (the third order tensor) by vectorization. However, such kind of vectorization will lose the original structure of the data and reduce the reliability of post processing, leading a poor representation. In this paper, we propose a new algorithm of overcomplete dictionary learning for tensor sparse coding, named K-CPD, by extension of K-SVD from vector formulation to tensor formulation. A multilinear orthogonal matching pursuit (MOMP) algorithm is also developed for calculating sparse representation of tensor signal. We evaluate the performance of K-CPD for image denoising, and the results demonstrate that the proposed method surpasses the conventional methods.
- 2012-05-10
著者
-
Chen Yen-wei
College Of Information And Science Ristumeikan University
-
DUAN Guifang
Global Innovation Research Institute, Ritsumeikan University
-
DENG Junping
College of Information Science and Engineering, Ritsumeikan University
関連論文
- A supervised LPP and neural network based scene classification with color histogram and camera metadata (パターン認識・メディア理解)
- A supervised LPP and neural network based scene classification with color histogram and camera metadata (コンピュータビジョンとイメージメディア)
- Facial pose image synthesis with a single image using tensor subspace approximation method (パターン認識・メディア理解)
- Facial pose image synthesis with a single image using tensor subspace approximation method (画像工学)
- Facial pose image synthesis with a single image using tensor subspace approximation method (医用画像)
- A novel automatic liver surface registration method using spherical harmonics (パターン認識・メディア理解)
- A novel automatic liver surface registration method using spherical harmonics (画像工学)
- A novel automatic liver surface registration method using spherical harmonics (医用画像)
- Multi-angle view, illumination and cosmetic facial image database (MaVIC) and quantitative analysis of facial appearance (パターン認識・メディア理解)
- Multi-angle view, illumination and cosmetic facial image database (MaVIC) and quantitative analysis of facial appearance (データ工学)
- Multi-class object recognition by fusion of image descriptors: classification evaluation of PASCAL VOC Challenge database (パターン認識・メディア理解)
- An adaptive color SIFT descriptor based on independent component analysis for image classification (パターン認識・メディア理解)
- D-7-17 Statistical Volume Modeling of Liver
- Segmentation of liver in low-contrast images using K-means clustering and a priori knowledge (パターン認識・メディア理解)
- Segmentation of liver in low-contrast images using K-means clustering and a priori knowledge (画像工学)
- Segmentation of liver in low-contrast images using K-means clustering and a priori knowledge (医用画像)
- D-16-2 ICA Based Ring Artifacts Reduction in Cone-Beam CT Images
- Reduction of ring artifacts in cone-beam CT images (医用画像)
- D-7-16 Parzen-Window Based Normalized Mutual Information for Image Registration
- 3D Image Registration and Its Application to Open-MR Scanner Based Liver Cancer Surgery(一般セッション(3),人の検出・計測・認識)
- 3D Image Registration and Its Application to Open-MR Scanner Based Liver Cancer Surgery(一般セッション(3),人の検出・計測・認識)
- 3D Image Registration and Its Application to Open-MR Scanner Based Liver Cancer Surgery
- A Wavelet-based Multiresolution Medical Image Registration Strategy Combining Mutual Information with Spatial Information
- Robust Edge Detection by Independent Component Analysis in Noisy Images(Image Processing and Video Processing)
- Joint Blind Super-Resolution and Shadow Removing(Image Processing and Video Processing)
- Comparison of Optimization Methods for Rigid Medical Image Registration
- D-16-10 Construction of Statistical Appearance Models from Fewer Samples by Generalized N-dimensional PCA
- Detection of the small bowel motility with cine-MRI sequence (医用画像)
- A semi-automated detection for motility of small bowel with MRI sequence (パターン認識・メディア理解)
- A semi-automated detection for motility of small bowel with MRI sequence (画像工学)
- A semi-automated detection for motility of small bowel with MRI sequence (医用画像)
- 3D Non-rigid Image Registration Assisted for Pneumoperitoneum in Laparoscopic Surgery
- Multilinear Supervised Neighborhood Embedding with Local Descriptor Tensor for Face Recognition
- Color Independent Components Based SIFT Descriptors for Object/Scene Classification
- Batch-incremental and robust principle component analysis (パターン認識・メディア理解)
- Batch-incremental and robust principle component analysis (マルチメディア・仮想環境基礎)
- Higher Order Orthogonal Iteration vs. Multilinear Principle Component Analysis for Tensor Subspace Learning
- Higher Order Orthogonal Iteration vs. Multilinear Principle Component Analysis for Tensor Subspace Learning
- Higher Order Orthogonal Iteration vs. Multilinear Principle Component Analysis for Tensor Subspace Learning
- Joint kernel equal integration of visual features and textual terms for biomedical imaging modality classification (医用画像)
- Independent components selection of color SIFT descriptors for image classification (医用画像)
- Joint kernel equal integration of visual features and textual terms for biomedical imaging modality classification (画像工学)
- Independent components selection of color SIFT descriptors for image classification (画像工学)
- Joint kernel equal integration of visual features and textual terms for biomedical imaging modality classification (パターン認識・メディア理解)
- Independent components selection of color SIFT descriptors for image classification (パターン認識・メディア理解)
- Higher Order Orthogonal Iteration vs. Multilinear Principle Component Analysis for Tensor Subspace Learning
- D-16-10 Multiple Normalized Statistic Shape Models and Components Selection for Assisting Cirrhosis Diagnosis
- Global Selection vs Local Ordering of Color SIFT Independent Components for Object/Scene Classification
- D-12-22 Visual Impression Estimation of Clothing Fabric Images Using Machine Learning Methods
- View-Based Object Recognition Using ND Tensor Supervised Neighborhood Embedding
- Multi-Linear Subspace Learning Methods for Statistical Texture Atlases of the Liver
- Fast and Robust Reconstruction of Penumbral Images by Combining Multiple Wiener Filters
- An Overcomplete Dictionary Learning Algorithm for Tensor Sparse Representation
- An Overcomplete Dictionary Learning Algorithm for Tensor Sparse Representation (画像工学)
- An Overcomplete Dictionary Learning Algorithm for Tensor Sparse Representation (医用画像)
- Multi-Linear Subspace Learning Methods for Statistical Texture Atlases of the Liver
- Multi-angle View, Illumination and Cosmetic Facial Image Database (MaVIC) and Quantitative Analysis of Facial Appearance
- Multi-angle View, Illumination and Cosmetic Facial Image Database (MaVIC) and Quantitative Analysis of Facial Appearance
- Segmentation of Liver in Low-Contrast Images Using K-Means Clustering and Geodesic Active Contour Algorithms
- 3D Non-rigid Medical Volume Registration using Landmark Based Initialization for Locoregional Therapy of Liver
- Multi-Linear Subspace Learning Methods for Statistical Texture Atlases of the Liver
- An Overcomplete Dictionary Learning Algorithm for Tensor Sparse Representation