Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Eigenvector-Based Distance Metric Learning for Image Classification and Retrieval

Eigenvector-Based Distance Metric Learning for Image Classification and Retrieval Distance metric learning has been widely studied in multifarious research fields. The mainstream approaches learn a Mahalanobis metric or learn a linear transformation. Recent related works propose learning a linear combination of base vectors to approximate the metric. In this way, fewer variables need to be determined, which is efficient when facing high-dimensional data. Nevertheless, such works obtain base vectors using additional data from related domains or randomly generate base vectors. However, obtaining base vectors from related domains requires extra time and additional data, and random vectors introduce randomness into the learning process, which requires sufficient random vectors to ensure the stability of the algorithm. Moreover, the random vectors cannot capture the rich information of the training data, leading to a degradation in performance. Considering these drawbacks, we propose a novel distance metric learning approach by introducing base vectors explicitly learned from training data. Given a specific task, we can make a sparse approximation of its objective function using the top eigenvalues and corresponding eigenvectors of a predefined integral operator on the reproducing kernel Hilbert space. Because the process of generating eigenvectors simply refers to the training data of the considered task, our proposed method does not require additional data and can reflect the intrinsic information of the input features. Furthermore, the explicitly learned eigenvectors do not result in randomness, and we can extend our method to any kernel space without changing the objective function. We only need to learn the coefficients of these eigenvectors, and the only hyperparameter that we need to determine is the number of eigenvectors that we utilize. Additionally, an optimization algorithm is proposed to efficiently solve this problem. Extensive experiments conducted on several datasets demonstrate the effectiveness of our proposed method. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) Association for Computing Machinery

Eigenvector-Based Distance Metric Learning for Image Classification and Retrieval

Download PDF
19 pages

Loading next page...
 
/lp/association-for-computing-machinery/eigenvector-based-distance-metric-learning-for-image-classification-HdLf3y70tZ

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Association for Computing Machinery
Copyright
Copyright © 2019 ACM
ISSN
1551-6857
eISSN
1551-6865
DOI
10.1145/3340262
Publisher site
See Article on Publisher Site

Abstract

Distance metric learning has been widely studied in multifarious research fields. The mainstream approaches learn a Mahalanobis metric or learn a linear transformation. Recent related works propose learning a linear combination of base vectors to approximate the metric. In this way, fewer variables need to be determined, which is efficient when facing high-dimensional data. Nevertheless, such works obtain base vectors using additional data from related domains or randomly generate base vectors. However, obtaining base vectors from related domains requires extra time and additional data, and random vectors introduce randomness into the learning process, which requires sufficient random vectors to ensure the stability of the algorithm. Moreover, the random vectors cannot capture the rich information of the training data, leading to a degradation in performance. Considering these drawbacks, we propose a novel distance metric learning approach by introducing base vectors explicitly learned from training data. Given a specific task, we can make a sparse approximation of its objective function using the top eigenvalues and corresponding eigenvectors of a predefined integral operator on the reproducing kernel Hilbert space. Because the process of generating eigenvectors simply refers to the training data of the considered task, our proposed method does not require additional data and can reflect the intrinsic information of the input features. Furthermore, the explicitly learned eigenvectors do not result in randomness, and we can extend our method to any kernel space without changing the objective function. We only need to learn the coefficients of these eigenvectors, and the only hyperparameter that we need to determine is the number of eigenvectors that we utilize. Additionally, an optimization algorithm is proposed to efficiently solve this problem. Extensive experiments conducted on several datasets demonstrate the effectiveness of our proposed method.

Journal

ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) – Association for Computing Machinery

Published: Aug 20, 2019

Keywords: Distance metric learning

References

  • Semi-supervised metric learning using pairwise constraints
    Baghshah, Mahdieh Soleymani; Shouraki, Saeed Bagheri
  • Semi-supervised metric learning using pairwise constraints
    Baghshah, Mahdieh Soleymani; Shouraki, Saeed Bagheri
  • Learning a mahalanobis metric from equivalence constraints
    Bar-Hillel, Aharon; Hertz, Tomer; Shental, Noam; Weinshall, Daphna
  • CP-mtML: Coupled projection multi-task metric learning for large scale face retrieval
    Bhattarai, Binod; Sharma, Gaurav; Jurie, Frédéric
  • Some faces are more equal than others: Hierarchical organization for accurate and efficient large-scale identity-based face retrieval
    Bhattarai, Binod; Sharma, Gaurav; Jurie, Frédéric; Pérez, Patrick
  • Distributed optimization and statistical learning via the alternating direction method of multipliers
    Boyd, Stephen; Parikh, Neal; Chu, Eric; Peleato, Borja; Eckstein, Jonathan
  • Information-theoretic metric learning
    Davis, Jason V.; Kulis, Brian; Jain, Prateek; Sra, Suvrit; Dhillon, Inderjit S.
  • Learning multiple kernel metrics for iterative person re-identification
    Dong, Husheng; Lu, Ping; Liu, Chunping; Ji, Yi; Gong, Shengrong
  • {n
    Fisher, R
  • Matrix computations
    Golub, Gene H.; Van Loan, Charles F.
  • An empirical feature-based learning algorithm producing sparse approximations
    Guo, Xin; Zhou, Ding-Xuan
  • Semi-supervised distance metric learning for collaborative image retrieval and clustering
    Hoi, Steven CH; Liu, Wei; Chang, Shih-Fu
  • Labeled Faces in the Wild: A Database for Studying Face Recognition in Unconstrained Environments
    Huang, Gary B.; Ramesh, Manu; Berg, Tamara; Learned-Miller, Erik
  • Regularized distance metric learning: Theory and algorithm
    Jin, Rong; Wang, Shijun; Zhou, Yang
  • Principal Component Analysis
    Jolliffe, Ian
  • The megaface benchmark: 1 million faces for recognition at scale
    Kemelmacher-Shlizerman, Ira; Seitz, Steven M.; Miller, Daniel; Brossard, Evan
  • Regularized large margin distance metric learning
    Li, Ya; Tian, Xinmei; Tao, Dacheng
  • Person re-identification by local maximal occurrence representation and metric learning
    Liao, Shengcai; Hu, Yang; Zhu, Xiangyu; Li, Stan Z
  • Constrained metric learning via distance gap maximization
    Liu, Wei; Tian, Xinmei; Tao, Dacheng; Liu, Jianzhuang
  • Decomposition-based transfer distance metric learning for image classification
    Luo, Yong; Liu, Tongliang; Tao, Dacheng; Xu, Chao
  • Transfer metric learning: Algorithms, applications and outlooks
    Luo, Yong; Wen, Yonggang; Duan, Lingyu; Tao, Dacheng
  • Transferring knowledge fragments for learning distance metric from a heterogeneous domain
    Luo, Yong; Wen, Yonggang; Liu, Tongliang; Tao, Dacheng
  • Pose-aware face recognition in the wild
    Masi, Iacopo; Rawls, Stephen; Medioni, Gérard; Natarajan, Prem
  • PCCA: A new approach for distance learning from sparse pairwise constraints
    Mignon, Alexis; Jurie, Frédéric
  • Boosted metric learning for efficient identity-based face retrieval
    Negrel, Romain; Lechervy, Alexis; Jurie, Frédéric
  • A data-driven approach to cleaning large face datasets
    Ng, Hong-Wei; Winkler, Stefan
  • Local fisher discriminant analysis for pedestrian re-identification
    Pedagadi, Sateesh; Orwell, James; Velastin, Sergio; Boghossian, Boghos
  • Think globally, fit locally: Unsupervised learning of low dimensional manifolds
    Saul, Lawrence K.; Roweis, Sam T.
  • Facenet: A unified embedding for face recognition and clustering
    Schroff, Florian; Kalenichenko, Dmitry; Philbin, James
  • Online and batch learning of pseudo-metrics
    Shalev-Shwartz, Shai; Singer, Yoram; Ng, Andrew Y.
  • User-click-data-based fine-grained image recognition via weakly supervised metric learning
    Tan, Min; Yu, Jun; Yu, Zhou; Gao, Fei; Rui, Yong; Tao, Dacheng
  • A global geometric framework for nonlinear dimensionality reduction
    Tenenbaum, Joshua B.; De Silva, Vin; Langford, John C.
  • Sparse transfer learning for interactive video search reranking
    Tian, Xinmei; Tao, Dacheng; Rui, Yong
  • Semi-supervised coefficient-based distance metric learning
    Wang, Zhangcheng; Li, Ya; Tian, Xinmei
  • Distance metric learning for large margin nearest neighbor classification
    Weinberger, Kilian Q.; Saul, Lawrence K.
  • Enhancing person re-identification in a self-trained subspace
    Yang, Xun; Wang, Meng; Hong, Richang; Tian, Qi; Rui, Yong
  • High-order distance based multiview stochastic learning in image classification
    Yu, Jun; Rui, Yong; Tang, Yuan Yan; Tao, Dacheng
  • Semantic preserving distance metric learning and applications
    Yu, Jun; Tao, Dapeng; Li, Jonathan; Cheng, Jun
  • Semisupervised multiview distance metric learning for cartoon synthesis
    Yu, Jun; Wang, Meng; Tao, Dacheng
  • Deep multimodal distance metric learning using click constraints for image ranking
    Yu, Jun; Yang, Xiaokang; Gao, Fei; Tao, Dacheng
  • Learning a discriminative null space for person re-identification
    Zhang, Li; Xiang, Tao; Gong, Shaogang
  • NESVM: A fast gradient method for support vector machines
    Zhou, Tianyi; Tao, Dacheng; Wu, Xindong
  • Kernel projection machine: A new tool for pattern recognition
    Zwald, Laurent; Blanchard, Gilles; Massart, Pascal; Vert, Régis