A tree-based dictionary learning framework
From MaRDI portal
Publication:5137935
binary partition treegeneralized adaptive Haar wavelet transformhierarchical clusteringK-meansmultiscale dictionary learningorthogonal matching pursuit
Nontrigonometric harmonic analysis involving wavelets and other special systems (42C40) Numerical aspects of computer graphics, image analysis, and computational geometry (65D18) Computing methodologies for image processing (68U10) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)
Abstract: We propose a new outline for adaptive dictionary learning methods for sparse encoding based on a hierarchical clustering of the training data. Through recursive application of a clustering method, the data is organized into a binary partition tree representing a multiscale structure. The dictionary atoms are defined adaptively based on the data clusters in the partition tree. This approach can be interpreted as a generalization of a discrete Haar wavelet transform. Furthermore, any prior knowledge on the wanted structure of the dictionary elements can be simply incorporated. The computational complexity of our proposed algorithm depends on the employed clustering method and on the chosen similarity measure between data points. Thanks to the multiscale properties of the partition tree, our dictionary is structured: when using Orthogonal Matching Pursuit to reconstruct patches from a natural image, dictionary atoms corresponding to nodes being closer to the root node in the tree have a tendency to be used with greater coefficients.
Recommendations
- Dictionary learning algorithms and applications
- Dictionary Learning Algorithms for Sparse Representation
- Dictionary Learning for L1-Exact Sparse Coding
- Direct Optimization of the Dictionary Learning Problem
- Efficient dictionary learning with sparseness-enforcing projections
- An MDL Framework for Sparse Coding and Dictionary Learning
- Compressed dictionary learning
- Dictionary Learning for Sparse Approximations With the Majorization Method
Cites work
- rm K-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation
- A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
- A first-order primal-dual algorithm for convex problems with applications to imaging
- A region-based easy-path wavelet transform for sparse image representation
- A wavelet tour of signal processing. The sparse way.
- Adaptive greedy approximations
- Analysis K-SVD: A Dictionary-Learning Algorithm for the Analysis Sparse Model
- Data-driven tight frame construction and image denoising
- Dictionary Pair Learning on Grassmann Manifolds for Image Denoising
- Greed is Good: Algorithmic Results for Sparse Approximation
- scientific article; zbMATH DE number 7195210 (Why is no real title available?)
- Iterative thresholding for sparse approximations
- Multi-Level Discriminative Dictionary Learning With Application to Large Scale Image Classification
- On Sparse Representation in Fourier and Local Bases
- Online learning for matrix factorization and sparse coding
- Proximal methods for hierarchical sparse coding
- Seismic data interpolation and denoising by learning a tensor tight frame
- Sparse and redundant representations. From theory to applications in signal and image processing.
- Sparse Approximate Solutions to Linear Systems
- The earth mover's distance as a metric for image retrieval
- The Haar wavelet transform of a dendrogram
- The mathematics of signal processing
- Trainlets: Dictionary Learning in High Dimensions
Cited in
(5)- Learning Decision Trees Using the Fourier Spectrum
- scientific article; zbMATH DE number 6129722 (Why is no real title available?)
- Efficiency of the weak Rescaled Pure Greedy Algorithm
- Proximal methods for hierarchical sparse coding
- Registration and inpainting of biological slice images based on Bendlet transform and interval Shannon-cosine wavelet
This page was built for publication: A tree-based dictionary learning framework
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5137935)
