Corpus ID: 16529932

A Multiscale Approach for Nonnegative Matrix Factorization with Applications to Image Classification

@inproceedings{Brezeale2012AMA,
  title={A Multiscale Approach for Nonnegative Matrix Factorization with Applications to Image Classification},
  author={Darin Brezeale},
  year={2012}
}
We use a multiscale approach to reduce the time to produce the nonnegative matrix factorization (NMF) of a matrix A, that is, A ≈ WH. We also investigate QR factorization as a method for initializing W during the iterative process for producing the nonnegative matrix factorization of A. Finally, we use our approach to produce nonnegative matrix factorizations for classifying images and compare it to the standard approach in terms of classification accuracy. 

References

SHOWING 1-10 OF 17 REFERENCES
Algorithms and applications for approximate nonnegative matrix factorization
TLDR
The development and use of low-rank approximate nonnegative matrix factorization algorithms for feature extraction and identification in the fields of text mining and spectral data analysis and the interpretability of NMF outputs in specific contexts are provided. Expand
Analyzing non-negative matrix factorization for image classification
TLDR
This paper experimentally compares NMF to Principal Component Analysis (PCA) in the context of image patch classification and finds that the two techniques are complementary and that their respective performance is correlated to the with-in class scatter. Expand
Incremental Nonnegative Matrix Factorization for Face Recognition
Nonnegative matrix factorization (NMF) is a promising approach for local feature extraction in face recognition tasks. However, there are two major drawbacks in almost all existing NMF-based methods.Expand
Email Surveillance Using Non-negative Matrix Factorization
TLDR
For the publicly released Enron electronic mail collection, sparse term-by-message matrices are encoded and a low rank non-negative matrix factorization algorithm is used to preserve natural data non-negativity and avoid subtractive basis vector and encoding interactions present in techniques such as principal component analysis. Expand
A Multi-Resolution Approach For Calculating Primary Eigenvectors Of a Large Set of Images
TLDR
A multi-resolution algorithm for calculating primary eigenvectors of a large set of high resolution images with substantial speedups over the often used SVD approach and is expected to run even faster as the underlying images' resolution gets higher. Expand
Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values†
A new variant ‘PMF’ of factor analysis is described. It is assumed that X is a matrix of observed data and σ is the known matrix of standard deviations of elements of X. Both X and σ are ofExpand
Learning the parts of objects by non-negative matrix factorization
TLDR
An algorithm for non-negative matrix factorization is demonstrated that is able to learn parts of faces and semantic features of text and is in contrast to other methods that learn holistic, not parts-based, representations. Expand
Matrix methods in data mining and pattern recognition
  • L. Eldén
  • Computer Science, Mathematics
  • Fundamentals of algorithms
  • 2007
TLDR
This book discusses vectors and matrices in data mining and pattern recognition, Linear Algebra Concepts and Matrix Decompositions, and Computing Eigenvalues and singular values. Expand
From Few to Many: Illumination Cone Models for Face Recognition under Variable Lighting and Pose
We present a generative appearance-based method for recognizing human faces under variation in lighting and viewpoint. Our method exploits the fact that the set of images of an object in fixed pose,Expand
UNDERSTANDING COMPLEX DATASETS: DATA MINING WITH MATRIX DECOMPOSITIONS
with the most likely topic assignments FIGURE 4.4 (SEE COLOR INSERT FOLLOWING PAGE 130.): The analysis of a document from Science. Document similarity was computed using Eq. (4.4); topic words wereExpand
...
1
2
...