Advances in subspace learning and its applications

• Bernardo B. Gatto UFAM
• Kazuhiro Fukui University of Tsukuba
• Waldir S. S. Júnior UFAM
• Eulanda M. dos Santos UFAM

DOI: https://doi.org/10.5753/sibgrapi.est.2021.20011

Resumo

Pattern-set matching refers to a class of problems where learning takes place through sets rather than elements. Much used in computer vision, this approach presents robustness to variations such as illumination, intrinsic parameters of the signal capture devices, and pose of the analyzed object. Inspired by applications of subspace analysis, three new collections of methods are presented in this thesis summary: (1) New representations for two-dimensional sets; (2) Shallow networks for image classification; and (3) Tensor data representation by subspaces. New representations are proposed to preserve the spatial structure and maintain a fast processing time. We also introduce a technique to keep temporal structure, even using the principal component analysis, which classically does not model sequences. In shallow networks, we present two convolutional neural networks that do not require backpropagation, employing only subspaces for their convolution filters. These networks present advantages when the training time and hardware resources are scarce. Finally, to handle tensor data, such as videos, we propose methods that employ subspaces for representation in a compact and discriminative way. Our proposed work has been applied in problems other than computer vision, such as representation and classification of bioacoustics and text patterns.

Referências

I. Jolliffe, “Principal component analysis,” in International encyclopedia of statistical science. Springer, 2011, pp. 1094–1096.

K.-i. Maeda, “From the subspace methods to the mutual subspace method,” in Computer Vision. Springer, 2010, pp. 135–156.

K. Fukunaga and W. L. Koontz, “Application of the karhunen-loeve expansion to feature selection and ordering,” IEEE Transactions on computers, vol. 100, no. 4, pp. 311–318, 1970.

J. Yang, D. Zhang, A. F. Frangi, and J.-y. Yang, “Two-dimensional pca: a new approach to appearance-based face representation and recognition,” Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 26, no. 1, pp. 131–137, 2004.

T.-H. Chan, K. Jia, S. Gao, J. Lu, Z. Zeng, and Y. Ma, “Pcanet: A simple deep learning baseline for image classification?” IEEE Transactions on Image Processing, vol. 24, no. 12, pp. 5017–5032, 2015.

K. Fukunaga, Introduction to statistical pattern recognition. Academic press, 2013.

Y. M. Lui, J. R. Beveridge, and M. Kirby, “Action classification on product manifolds,” in Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on. IEEE, 2010, pp. 833–839.

G. H. Golub et al., “Cf van loan, matrix computations,” The Johns Hopkins, 1996.

K. Fukui and A. Maki, “Difference subspace and its generalization for subspace-based methods,” IEEE transactions on pattern analysis and machine intelligence, vol. 37, no. 11, pp. 2164–2177, 2015.

B. B. Gatto, E. M. dos Santos, A. L. Koerich, K. Fukui, and W. S. Junior, “Tensor analysis with n-mode generalized difference subspace,” Expert Systems with Applications, vol. 171, p. 114559, 2021.

L. S. Souza, B. B. Gatto, and K. Fukui, “Classification of bioacoustic signals with tangent singular spectrum analysis,” in ICASSP 2019- 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2019, pp. 351–355.

——, “Grassmann singular spectrum analysis for bioacoustics classification,” in 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2018, pp. 256–260.

B. B. Gatto, E. M. dos Santos, J. G. Colonna, N. Sogi, L. S. Souza, and K. Fukui, “Discriminative singular spectrum analysis for bioacoustic classification,” in INTERSPEECH 2020. International Speech Communication Association (ISCA), 2020.

B. B. Gatto and E. M. dos Santos, “Discriminative canonical correlation analysis network for image classification,” in 2017 IEEE International Conference on Image Processing (ICIP). IEEE, 2017, pp. 4487–4491.

L. S. Souza, N. Sogi, B. B. Gatto, T. Kobayashi, and K. Fukui, “An interface between grassmann manifolds and vector spaces,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020, pp. 846–847.

N. Sogi, L. S. Souza, B. B. Gatto, and K. Fukui, “Metric learning with a-based scalar product for image-set recognition,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020, pp. 850–851.

B. B. Gatto, E. M. dos Santos, and K. Fukui, “Subspace-based convolutional network for handwritten character recognition,” in Document Analysis and Recognition (ICDAR), 2017 14th IAPR International Conference on, vol. 1. IEEE, 2017, pp. 1044–1049.

E. K. Shimomoto, L. S. Souza, B. B. Gatto, and K. Fukui, “Text classification based on word subspace with term-frequency,” in 2018 International Joint Conference on Neural Networks (IJCNN). IEEE, 2018, pp. 1–8.

B. B. Gatto, W. S. da Silva, and E. M. dos Santos, “Kernel two dimensional subspace for image set classification,” in IEEE International Conference on Tools with Artificial Intelligence (ICTAI). IEEE, 2016, pp. 1004–1011.

B. B. Gatto, E. M. dos Santos, and W. S. da Silva, “Orthogonal hankel subspaces for applications in gesture recognition,” in 2017 30th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI). IEEE, 2017, pp. 429–435.

B. B. Gatto, A. Bogdanova, L. S. Souza, and E. M. dos Santos, “Hankel subspace method for efficient gesture representation,” in 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP). IEEE, 2017, pp. 1–6.

B. B. Gatto, J. G. Colonna, E. M. dos Santos, and E. F. Nakamura, “Mutual singular spectrum analysis for bioacoustics classification,” in 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP). IEEE, 2017, pp. 1–6.

B. B. Gatto and E. M. Dos Santos, “Image-set matching by two dimensional generalized mutual subspace method,” in Brazilian Conference on Intelligent Systems (BRACIS). IEEE, 2016, pp. 133–138.

B. B. Gatto, M. A. Molinetti, E. M. dos Santos, and K. Fukui, “Tensor fukunaga-koontz transform for hierarchical clustering,” in 2019 8th Brazilian Conference on Intelligent Systems (BRACIS). IEEE, 2019, pp. 150–155.

B. B. Gatto, L. S. de Souza, and E. M. dos Santos, “A deep network model based on subspaces: A novel approach for image classification,” in Machine Vision Applications (MVA), 2017 Fifteenth IAPR International Conference on. IEEE, 2017, pp. 436–439.

L. S. de Souza, B. B. Gatto, and K. Fukui, “Enhancing discriminability of randomized time warping for motion recognition,” in 2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA). IEEE, 2017, pp. 77–80.

E. K. Shimomoto, L. S. Souza, B. B. Gatto, and K. Fukui, “News2meme: An automatic content generator from news based on word subspaces from text and image,” in 2019 16th International Conference on Machine Vision Applications (MVA). IEEE, 2019, pp. 1–6.

B. B. Gatto, E. M. dos Santos, M. A. Molinetti, and K. Fukui, “Multilinear clustering via tensor fukunaga-koontz transform with fisher eigenspectrum regularization,” Applied Soft Computing, 2021.

L. S. Souza, B. B. Gatto, J.-H. Xue, and K. Fukui, “Enhanced grassmann discriminant analysis with randomized time warping for motion recognition,” Pattern Recognition, vol. 97, p. 107028, 2020.

B. B. Gatto, E. M. dos Santos, K. Fukui, W. S. Júnior, and K. V. dos Santos, “Fukunaga–koontz convolutional network with applications on character classification,” Neural Processing Letters, vol. 52, pp. 443– 465, 2020.

B. B. Gatto, L. S. Souza, E. M. dos Santos, K. Fukui, W. S. Júnior, and K. V. dos Santos, “A semi-supervised convolutional neural network based on subspace representation for image classification,” EURASIP Journal on Image and Video Processing, vol. 2020, no. 1, pp. 1–21, 2020.