Coupled Marginal Fisher Analysis for Low-Resolution Face Recognition

Siena, Stephen; Boddeti, Vishnu Naresh; Kumar, B. V. K. Vijaya

doi:10.1007/978-3-642-33868-7_24

Cited by 29 publications

(22 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Five cross-domain learning methods (DAMA [32], CCA [10], CMFA [29], MMCM [30] and DTRSVM [20]), which can cope with different scale domains, were also applied to LR re-id. However, none of them was designed for L-R re-id.…”

Section: Comparison With Cross-domain Methodsmentioning

confidence: 99%

“…In the last five years there are several coupled transformation based subspace models developed for LR face recognition [30,29,2,24,43,14]. A basic idea of these works is to learn coupled transformations such that a LR image can directly match a HR image.…”

Section: Related Workmentioning

confidence: 99%

“…Since none of the existing re-id approaches explicitly addresses the LR person re-id problem as we do, we further compared six other relevant methods. Specifically, five cross-domain learning methods were compared, including domain adaptation based on manifold alignment (DAMA) [32], canonical correlation analysis (CCA) [10], coupled marginal fisher analysis (CMFA) [29], maximummargin coupled mappings (MMCM) [30] and the DTRSVM in [20]. Among the five methods mentioned above, DAMA is for general purpose, DTRSVM is for domain adaptation in person re-identification but not for LR image processing, and CCA, CMFA and MMCM are representative methods for LR face image matching.…”

Section: Datasets and Settingsmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Scale Learning for Low-Resolution Person Re-Identification

Zheng

Wang

et al. 2015

2015 IEEE International Conference on Computer Vision (ICCV)

142

View full text Add to dashboard Cite

In real world person re-identification (re-id), images of people captured at very different resolutions from different locations need be matched. Existing re-id models typically normalise all person images to the same size. However, a low-resolution (LR) image contains much less information about a person, and direct image scaling and simple size normalisation as done in conventional re-id methods cannot compensate for the loss of information. To solve this LR person re-id problem, we propose a novel joint multi-scale learning framework, termed joint multi-scale discriminant component analysis (JUDEA). The key component of this framework is a heterogeneous class mean discrepancy (HCMD) criterion for cross-scale image domain alignment, which is optimised simultaneously with discriminant modelling across multiple scales in the joint learning framework. Our experiments show that the proposed JUDEA framework outperforms existing representative reid methods as well as other related LR visual matching models applied for the LR person re-id problem.

show abstract

Section: Comparison With Cross-domain Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Datasets and Settingsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Scale Learning for Low-Resolution Person Re-Identification

Zheng

Wang

et al. 2015

2015 IEEE International Conference on Computer Vision (ICCV)

142

View full text Add to dashboard Cite

show abstract

“…Zhou et al [15] designed coupled linear mappings based on the idea of classical discriminant analysis. Siena et al [16] obtained coupled mappings by extending the principles of Marginal Fisher Analysis (MFA) [17], which is an effective linear projection method for images in the same resolution. Biswas et al [18] improved the matching performance of the LR face images based on Multidimensional Scaling (MDS).…”

Section: Introductionmentioning

confidence: 99%

From Local Geometry to Global Structure: Learning Latent Subspace for Low-resolution Face Image Recognition

Shi

2015

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

In this letter, we propose a novel approach for learning coupled mappings to improve the performance of low-resolution (LR) face image recognition. The coupled mappings aim to project the LR probe images and high-resolution (HR) gallery images into a unified latent subspace, which is efficient to measure the similarity of face images with different resolutions. In the training phase, we first construct local optimization for each training sample according to the relationship of neighboring data points. The local optimization aims to: (1) ensure the consistency for each LR face image and corresponding HR one; (2) model the intrinsic geometric structure between each given sample and its neighbors; and (3) preserve the discriminative information across different subjects. We finally incorporate the local optimizations together for building the global structure. The coupled mappings can be learned by solving a standard eigen-decomposition problem, which avoids the small-sample-size problem. Experimental results demonstrate the effectiveness of the proposed method on public face databases.

show abstract

“…To allow for generalized eigenvalue decomposition solutions the model parameters and variables are treated as concatenated matrices. Methods in this category include Coupled Locality Preserving Mappings [7], Piecewise Regularized Canonical Correlation Discrimination [9], Simultaneous Discriminant Analysis (SDA) [18], Coupled Marginal Fisher Analysis (CMFA) [10], and Supervised Locality Preserving Projection Coupled Metric Learning [19]. The differences of these approaches lie in the way that they define the components of the objective function.…”

Section: Introductionmentioning

confidence: 99%

Semi-coupled basis and distance metric learning for cross-domain matching: Application to low-resolution face recognition

Moutafis

Kakadiaris

2014

IEEE International Joint Conference on Biometrics

View full text Add to dashboard Cite

In this paper, we propose a method for matching biometric data from disparate domains. Specifically, we focus on the problem of comparing a low-resolution (LR) image with a high-resolution (HR) one. Existing coupled mapping methods do not fully exploit the HR information or they do not simultaneously use samples from both domains during training. To this end, we propose a method that learns coupled distance metrics in two steps. In addition, we propose to jointly learn two semi-coupled bases that yield optimal representations. In particular, the HR images are used to learn a basis and a distance metric that result in increased class-separation. The LR images are then used to learn a basis and a distance metric that map the LR data to their class-discriminated HR pairs. Finally, the two distance metrics are refined to simultaneously enhance the class-separation of both HR class-discriminated and LR projected images. We illustrate that different distance metric learning approaches can be employed in conjunction with our framework. Experimental results on Multi-PIE and SCface, along with the relevant hypothesis tests, provide evidence of the effectiveness of the proposed approach. Figure 4. Depiction of the ROC curves for the Multi-PIE database for Experiment 2 (color figure, best viewed in electronic format).

show abstract

Coupled Marginal Fisher Analysis for Low-Resolution Face Recognition

Cited by 29 publications

References 9 publications

Multi-Scale Learning for Low-Resolution Person Re-Identification

Multi-Scale Learning for Low-Resolution Person Re-Identification

From Local Geometry to Global Structure: Learning Latent Subspace for Low-resolution Face Image Recognition

Semi-coupled basis and distance metric learning for cross-domain matching: Application to low-resolution face recognition

Contact Info

Product

Resources

About