Sparse image coding using a 3D non-negative tensor factorization
Authors:
- 1 January 2005
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1 (15505499) , 50-57 Vol. 1
Abstract
We introduce an algorithm for a non-negative 3D tensor factorization for the purpose of establishing a local parts feature decomposition from an object class of images. In the past, such a decomposition was obtained using non-negative matrix factorization (NMF) where images were vectorized before being factored by NMF. A tensor factorization (NTF) on the other hand preserves the 2D representations of images and provides a unique factorization (unlike NMF which is not unique). The resulting "factors" from the NTF factorization are both sparse (like with NMF) but also separable allowing efficient convolution with the test image. Results show a superior decomposition to what an NMF can provide on all fronts - degree of sparsity, lack of ghost residue due to invariant parts and efficiency of coding of around an order of magnitude better. Experiments on using the local parts decomposition for face detection using SVM and Adaboost classifiers demonstrate that the recovered features are discriminatory and highly effective for classification.Keywords
This publication has 20 references indexed in Scilit:
- Linear image coding for regression and classification using the tensor-rank principle2005
- Non-negative tensor factorization with applications to statistics and computer vision2005
- Real time pattern matching using projection kernels2003
- Learning and recognition of 3D objects from appearance2002
- A Multilinear Singular Value DecompositionSIAM Journal on Matrix Analysis and Applications, 2000
- Learning the parts of objects by non-negative matrix factorizationNature, 1999
- Emergence of simple-cell receptive field properties by learning a sparse code for natural imagesNature, 1996
- Positive matrix factorization: A non‐negative factor model with optimal utilization of error estimates of data valuesEnvironmetrics, 1994
- Low-dimensional procedure for the characterization of human facesJournal of the Optical Society of America A, 1987
- Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statisticsLinear Algebra and its Applications, 1977