MPCA: Multilinear Principal Component Analysis of Tensor Objects

@article{Lu2008MPCAMP,
  title={MPCA: Multilinear Principal Component Analysis of Tensor Objects},
  author={Haiping Lu and Konstantinos N. Plataniotis and Anastasios N. Venetsanopoulos},
  journal={IEEE Transactions on Neural Networks},
  year={2008},
  volume={19},
  pages={18-39}
}
This paper introduces a multilinear principal component analysis (MPCA) framework for tensor object feature extraction. Objects of interest in many computer vision and pattern recognition applications, such as 2D/3D images and video sequences are naturally described as tensors or multilinear arrays. The proposed framework performs feature extraction by determining a multilinear projection that captures most of the original tensorial input variation. The solution is iterative in nature and it… 
Uncorrelated Multilinear Principal Component Analysis for Unsupervised Multilinear Subspace Learning
TLDR
Experimental results included in this paper suggest that UMPCA is particularly effective in determining the low-dimensional projection space needed in such recognition tasks, and offers a way to systematically determine the maximum number of uncorrelated multilinear features that can be extracted by the method.
MPCA+MDA: A novel approach for face recognition based on tensor objects
TLDR
A novel approach to solve the supervised dimensionality reduction problem and feature extraction by encoding an image object as a general tensor of 2-D/3-D order and it is shown that MPCA+MDA algorithm based on higher order tensors has the potential to outperform the traditional vector-based subspace learning algorithms such as Eigenface and Fisherface.
Multilinear Sparse Principal Component Analysis
TLDR
The key operation of MSPCA is to rewrite the MPCA into multilinear regression forms and relax it for sparse regression, which has the potential to outperform the existing PCA-based subspace learning algorithms.
Multilinear Principal Component Analysis Network for Tensor Object Classification
TLDR
A multilinear principal component analysis network (MPCANet), which is a tensor extension of PCANet, to extract the high-level semantic features from multi-dimensional images, and it is shown that even a simple one-layer MPCANet may outperform a two-layer PCANets.
Uncorrelated Multilinear Discriminant Analysis With Regularization and Aggregation for Tensor Object Recognition
TLDR
The UMLDA aims to extract uncorrelated discriminative features directly from tensorial data through solving a tensor-to-vector projection, and an adaptive regularization procedure is incorporated to enhance the performance in the small sample size (SSS) scenario.
Ranking Tensor Subspaces in Weighted Multilinear Principal Component Analysis
TLDR
A method to explicitly incorporate prior information semantics in the MPCA framework to allow an automatic selective treatment of the variables that compose the patterns of interest and rely on spatial weights calculated by separating hyperplanes and Fisher criterion.
Tensorial Kernel Principal Component Analysis for Action Recognition
TLDR
This method aims to remedy the shortcomings of multilinear subspace learning (tensorial PCA) developed recently in modelling the nonlinear manifold of tensor objects and brings together the desirable properties of kernel methods and tensor decompositions for significant performance gain when the data are multidimensional and nonlinear dependencies do exist.
Comparing Ranking Methods for Tensor Components in Multilinear and Concurrent Subspace Analysis with Applications in Face Images
TLDR
If CSA receives centered input samples and it considers full projection matrices then the obtained solution is equal to the one generated by MPCA, and the general problem of ranking tensor components is considered.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 48 REFERENCES
Multilinear Principal Component Analysis of Tensor Objects for Recognition
TLDR
Experiments show that the proposed MPCA performs better than the baseline algorithm in human identification on the gait challenge data sets and the analogous counterparts in MPCA to the eigenvalues and eigenvectors in PCA are defined.
Discriminant analysis with tensor representation
TLDR
This paper proposes a discriminant tensor criterion (DTC), whereby multiple interrelated lower-dimensional discriminative subspaces are derived for feature selection and an algorithm discriminant analysis with tensor representation (DATER), which has the potential to outperform the traditional subspace learning algorithms, especially in the small sample size cases.
Tensor Subspace Analysis
TLDR
A new algorithm called Tensor Subspace Analysis (TSA) is proposed that detects the intrinsic local geometrical structure of the tensor space by learning a lower dimensional tensor subspace and achieves better recognition rate, while being much more efficient.
Multilinear Discriminant Analysis for Face Recognition
TLDR
This paper presents a novel approach to solve the supervised dimensionality reduction problem by encoding an image object as a general tensor of second or even higher order, and proposes a discriminant tensor criterion, whereby multiple interrelated lower dimensional discriminative subspaces are derived for feature extraction.
Concurrent subspaces analysis
TLDR
A new algorithm, called concurrent subspaces analysis (CSA), to derive representative subspace by encoding image objects as 2/sup nd/ or even higher order tensors, which outperforms principal component analysis in terms of both reconstruction and classification capability.
Multilinear Analysis of Image Ensembles: TensorFaces
TLDR
This work considers the multilinear analysis of ensembles of facial images that combine several modes, including different facial geometries (people), expressions, head poses, and lighting conditions, and concludes that the resulting "TensorFaces" representation has several advantages over conventional eigenfaces.
Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection
TLDR
A face recognition algorithm which is insensitive to large variation in lighting direction and facial expression is developed, based on Fisher's linear discriminant and produces well separated classes in a low-dimensional subspace, even under severe variations in lighting and facial expressions.
A neural-network appearance-based 3-D object recognition using independent component analysis
TLDR
This paper presents results on appearance-based three-dimensional (3-D) object recognition (3DOR) accomplished by utilizing a neural-network architecture developed based on independent component analysis (ICA), suggesting that the use of ICA may not necessarily always give better results than PCA, and that the application of I CA is highly data dependent.
Human Gait Recognition With Matrix Representation
TLDR
This matrix-based scheme demonstrates a much better gait recognition performance than state-of-the-art algorithms on the standard USF HumanID Gait database.
Two-dimensional PCA: a new approach to appearance-based face representation and recognition
TLDR
A new technique coined two-dimensional principal component analysis (2DPCA) is developed for image representation that is based on 2D image matrices rather than 1D vectors so the image matrix does not need to be transformed into a vector prior to feature extraction.
...
1
2
3
4
5
...