Metric Learning with A-based Scalar Product for Image-set Recognition

Sogi, Naoya; Souza, Lincon S.; Gatto, Bernardo B.; Fukui, Kiichi

doi:10.1109/cvprw50498.2020.00433

Cited by 10 publications

(3 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A popular procedure for measuring the similarity between subspaces is by computing the principal angles, also known as canonical angles [7,8]. Jordan introduced the theory of computing the canonical angles between subspaces.…”

Section: Computing the Similarity Between Subspacesmentioning

confidence: 99%

Pattern-set Representations using Linear, Shallow and Tensor Subspaces

Gatto

Santos²,

Júnior³

2022

CLEIej

View full text Add to dashboard Cite

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 paper: (1) New representations for two-dimensional sets; (2) Shallow networks for image classification; and (3) Subspaces for tensor representation and classification. New representations are proposed with the aim of preserving the spatial structure and maintaining 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 its convolution filters. These networks present advantages when the training time and hardware resources are scarce. Finally, to handle tensor data, such as video data, we propose methods that employ subspaces for representation in a compact and discriminative way. Our proposed work has been applied to several problems, such as 2D data representation, shallow networks for image classification, and tensor representation and learning.

show abstract

Section: Computing the Similarity Between Subspacesmentioning

confidence: 99%

Pattern-set Representations using Linear, Shallow and Tensor Subspaces

Gatto

Santos²,

Júnior³

2022

CLEIej

View full text Add to dashboard Cite

show abstract

“…Sogi et al [52] propose a metric learning approach for image set recognition. The main task is to provide a reliable metric for subspace representation.…”

Section: Subspace-based Learningmentioning

confidence: 99%

Grassmannian learning mutual subspace method for image set recognition

Souza¹,

Sogi²,

Gatto³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

This paper addresses the problem of object recognition given a set of images as input (e.g., multiple camera sources and video frames). Convolutional neural network (CNN)-based frameworks do not exploit these sets effectively, processing a pattern as observed, not capturing the underlying feature distribution as it does not consider the variance of images in the set. To address this issue, we propose the Grassmannian learning mutual subspace method (G-LMSM), a NN layer embedded on top of CNNs as a classifier, that can process image sets more effectively and can be trained in an end-to-end manner. The image set is represented by a low-dimensional input subspace; and this input subspace is matched with reference subspaces by a similarity of their canonical angles, an interpretable and easy to compute metric. The key idea of G-LMSM is that the reference subspaces are learned as points on the Grassmann manifold, optimized with Riemannian stochastic gradient descent. This learning is stable, efficient and theoretically well-grounded. We demonstrate the effectiveness of our proposed method on hand shape recognition, face identification, and facial emotion recognition.

show abstract

“…Some work also received international awards. A1 ICIP [14] A1 CVPRW [15], [16] A1 ICDAR [17] A2 IJCNN [18] A2 ICTAI [19] A3 SIBGRAPI [20] A3 MLSP [21], [22] A4 BRACIS [23], [24] A4 MVA [25], [26], [27] B1 A1 ASOC [28] A1 PR [29] A1 NEPL [30] A3 EURASIP JIVP [31] A4…”

Section: Awards Publications and Distinctionsmentioning

confidence: 99%

Advances in subspace learning and its applications

Gatto¹,

Fukui²,

Júnior³

et al. 2021

Anais Estendidos Da XXXIV Conference on Graphics, Patterns and Images (SIBRAPI Estendido 2021)

View full text Add to dashboard Cite

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.

show abstract

Metric Learning with A-based Scalar Product for Image-set Recognition

Cited by 10 publications

References 39 publications

Pattern-set Representations using Linear, Shallow and Tensor Subspaces

Pattern-set Representations using Linear, Shallow and Tensor Subspaces

Grassmannian learning mutual subspace method for image set recognition

Advances in subspace learning and its applications

Contact Info

Product

Resources

About