SPD Data Dimensionality Reduction based on SPD Manifold Tangent Space and Local LDA
Pages 68 - 73
Abstract
The bilinear transformation of symmetric positive definite (SPD) matrix is currently the most common SPD data dimensionality reduction framework. The transformation matrix can be determined according to different criteria to produce different SPD data dimensionality reduction algorithms. The bilinear transformation of SPD matrix is essentially a transformation between two Riemannian manifolds. Since Riemannian manifolds are not linear spaces, most SPD data bilinear transformation dimensionality reduction algorithms now avoid linear operations between SPD matrices. This paper proposes a SPD data dimensionality reduction algorithm based on SPD manifold tangent space and local LDA. The main contributions are as follows: (1) This paper adopts the affine invariant Riemann metric, the tangent space of the unit matrix and the log transformation to transform the SPD The data is transformed from the Riemannian manifold to the unit matrix tangent space, so that the geodesic distance between the SPD and the unit matrix and the Euclidean distance transformation between the tangent vector and the origin of the tangent space after the transformation remain unchanged; (2) The so-called local discriminant difference of SPD is to first The SPD data is decomposed into one part, and then the local discriminant difference of the SPD data is calculated according to the category (label) of the SPD data. The local discriminant difference of SPD data better realizes the combination of internal and external attributes of SPD data.
References
[1]
Jayasumana S, Hartley R, Salzmann M, li H, Harandi M (2014) Kernel Methods on Riemannian Manifolds with Gaussian RBF Kernels. IEEE Transactions on Pattern Analysis and Machine Intelligence 37.
[2]
Huang Z, Wang R, Shan S, Gool LV, Chen X (2018) Cross Euclidean-to-Riemannian Metric Learning with Application to Face Recognition from Video. IEEE Transactions on Pattern Analysis and Machine Intelligence 40 (12):2827-2840. 2017.2776154.
[3]
Wang H, Wang Q, Gao M, Li P, Zuo W Multi-scale Location-Aware Kernel Representation for Object Detection. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 18-23 June 2018 2018. pp 1248-1257.
[4]
Li P, Wang Q, Zuo W, Zhang L Log-Euclidean Kernels for Sparse Representation and Dictionary Learning. In: 2013 IEEE International Conference on Computer Vision, 1-8 Dec. 2013 2013. pp 1601-1608.
[5]
Zhang J, Zhou L, Wang L, Li W (2015) Functional Brain Network Classification With Compact Representation of SICE Matrices. IEEE Transactions on Biomedical Engineering 62 (6):1623-1634.
[6]
Huang Z, Wang R, Li X, Liu W, Shan S, Gool LV, Chen X (2018) Geometry-Aware Similarity Learning on SPD Manifolds for Visual Recognition. IEEE Transactions on Circuits and Systems for Video Technology 28 (10):2513-2523.
[7]
Ren J, Wu XJ (2020) Probability Distribution-Based Dimensionality Reduction on Riemannian Manifold of SPD Matrices. IEEE Access 8:153881-153890.
[8]
Xu C, Lu C, Gao J, Zheng W, Wang T, Yan S (2015) Discriminative Analysis for Symmetric Positive Definite Matrices on Lie Groups. IEEE Transactions on Circuits and Systems for Video Technology 25 (10):1576-1585.
[9]
Sanin A, Sanderson C, Harandi MT, Lovell BC Spatio-temporal covariance descriptors for action and gesture recognition. In: 2013 IEEE Workshop on Applications of Computer Vision (WACV), 15-17 Jan. 2013 2013. pp 103-110.
[10]
Wang R, Guo H, Davis LS, Dai Q Covariance discriminative learning: A natural and efficient approach to image set classification. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition, 16-21 June 2012 2012. pp 2496-2503.
[11]
Hussein M, Torki M, Gowayyed M, El Saban M (2013) Human Action Recognition Using a Temporal Hierarchy of Covariance Descriptors on 3D Joint Locations.
[12]
Guo K, Ishwar P, Konrad J (2013) Action Recognition From Video Using Feature Covariance Matrices. IEEE Transactions on Image Processing 22 (6):2479-2494.
[13]
Harandi MT, Salzmann M, Hartley R From Manifold to Manifold: Geometry-Aware Dimensionality Reduction for SPD Matrices. In: Fleet D, Pajdla T, Schiele B, Tuytelaars T (eds) Computer Vision – ECCV 2014, Cham, 2014// 2014. Springer International Publishing, pp 17-32
[14]
Huang Z, Gool LV (2016) A Riemannian Network for SPD Matrix Learning. CoRR abs/1608.04233.
[15]
M. Müller, T. Röder, M. Clausen, B. Eberhardt, B. Krüger, and A. Weber, “Documentation: Mocap database HDM05,” Universität Bonn, Bonn, Germany, Tech. Rep. CG-2007-2, 2007
[16]
Wang W, Wang R, Huang Z, Shan S, Chen X (2018) Discriminant Analysis on Riemannian Manifold of Gaussian Distributions for Face Recognition With Image Sets. IEEE Transactions on Image Processing 27 (1):151-163.
[17]
Xie X, Yu Z, Gu Z, Li Y (2018) Classification of Symmetric Positive Definite Matrices based on Bilinear Isometric Riemannian Embedding. Pattern Recognition 87.
[18]
Harandi M, Salzmann M, Hartley R (2018) Dimensionality Reduction on SPD Manifolds: The Emergence of Geometry-Aware Methods. IEEE Transactions on Pattern Analysis and Machine Intelligence 40 (1):48-62.
[19]
Santi Phithakkitnukoon and Carlo Ratti, "A Recent-Pattern Biased Dimension-Reduction Framework for Time Series Data," Journal of Advances in Information Technology, Vol. 1, No. 4, pp. 168-180, November, 2010.
[20]
Arpita Joshi and Nurit Haspel, "Clustering of Protein Conformations Using Parallelized Dimensionality Reduction," Journal of Advances in Information Technology, Vol. 10, No. 4, pp. 142-147, November 2019.
Recommendations
Dimensionality reduction of SPD data based on Riemannian manifold tangent spaces and local affinity
AbstractNon-Euclidean data is increasingly used in practical applications. As a typical representative, Symmetric Positive Definite (SPD) matrices can form a Riemannian manifold rather than a flat linear space. Hence, constructing a dimensionality ...
Local Tangent Space Discriminant Analysis
We propose a novel supervised dimensionality reduction method named local tangent space discriminant analysis (TSD) which is capable of utilizing the geometrical information from tangent spaces. The proposed method aims to seek an embedding space where ...
Tangent-based manifold approximation with locally linear models
In this paper, we consider the problem of manifold approximation with affine subspaces. Our objective is to discover a set of low dimensional affine subspaces that represent manifold data accurately while preserving the manifold@?s structure. For this ...
Comments
Information & Contributors
Information
Published In
September 2021
139 pages
ISBN:9781450385084
DOI:10.1145/3490700
Copyright © 2021 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 05 January 2022
Check for updates
Author Tags
Qualifiers
- Research-article
- Research
- Refereed limited
Funding Sources
- ??5G?????????????????????
Conference
ICACS '21
ICACS '21: 2021 The 5th International Conference on Algorithms, Computing and Systems
September 24 - 26, 2021
Xi'an, China
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 69Total Downloads
- Downloads (Last 12 months)8
- Downloads (Last 6 weeks)0
Reflects downloads up to 09 Nov 2024
Other Metrics
Citations
View Options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign inFull Access
View options
View or Download as a PDF file.
PDFeReader
View online with eReader.
eReaderHTML Format
View this article in HTML Format.
HTML Format