Face Recognition with the Multiple Constrained Mutual Subspace Method

Nishiyama, Masashi; Yamaguchi, Osamu; Fukui, Kiichi

doi:10.1007/11527923_8

Cited by 54 publications

(35 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PCA-based method is known as Eigenface 12) and LDA-based method is known as Fisherface 13) . Other methods employ subspace methods such as CLAss-Featuring Information Compression (CLAFIC) method 14) , subspace method 14) , mutual subspace method and its extensions such as constrained mutual subspace method 15) , and multiple constrained mutual subspace method 16) , etc., where such a feature is represented as a set of bases of the subspace. The approaches mentioned above transform the highdimensional image space into the low-dimensional subspaces and provide good representation and good dis-crimination for face recognition by selecting effective subspaces.…”

Section: Feature Extraction and Matchingmentioning

confidence: 99%

[Invited Paper] Recent Advances in Biometric Recognition

Ito

Aoki

2018

MTA

View full text Add to dashboard Cite

This paper presents recent advances in biometric recognition, where we focus on face, fingerprint and iris recognition, which are major research topics on biometric recognition. We summarize the research trend of face, fingerprint and iris recognition over the past decade. This paper also presents our activities of biometric recognition. Our approach employs the phase information obtained by Discrete Fourier Transform (DFT) of images. The phase information preserves the inherent features of the image, and its correlation function, called phase correlation or Phase-Only Correlation (POC), gives us both the good similarity measure for biometric recognition and the translational displacement for image registration.Our approach of using phase information has been successfully applied to fingerprint, face, iris, palmprint, finger knuckle and dental recognition. Among them, we present some interesting results of palmprint recognition, finger knuckle recognition and dental recognition.

show abstract

Section: Feature Extraction and Matchingmentioning

confidence: 99%

[Invited Paper] Recent Advances in Biometric Recognition

Ito

Aoki

2018

MTA

View full text Add to dashboard Cite

show abstract

“…Method Key-frame based Approaches [90], [40], [47], [114], [100], [17], [115], [31], [78], [85], [98], [101], [118] Temporal Model based Approaches [74], [73], [72], [75], [18], [24], [67], [69], [68], [122], [120], [123], [121], [64], [65], [66], [79], [55], [2], [43], [50], [49] Image-Set Matching based Approaches Statistical model-based [93], [4], [96], [7], [10], [6], [9] Mutual subspace-based [110], [90], [35], [82], [108], [56], [57], [5]<...>…”

Section: Categorymentioning

confidence: 99%

“…This is based on the assumption that images are drawn from some distributions on the underlying face pattern manifold, and normally statistical learning algorithms are adopted to model the distribution. Recently, following the mutual subspace method [110], many approaches build a compact model of the distribution by representing each image set as a linear subspace, and measure their similarity using the canonical angles [110,82,59]. In the following sections, we discuss these two groups of approaches: statistical model-based and mutual subspace-based, respectively.…”

Section: Image-set Matching Based Approachesmentioning

confidence: 99%

“…A real-time system implemented using CMSM was demonstrated in [62]. In [82], the authors further introduced the Multiple Constrained Mutual Subspace Method (MCMSM), which generates multiple constraint subspaces by using the ensemble learning algorithms (Bagging and Boosting). The similarities on each constraint subspace are combined for recognition.…”

Section: Mutual Subspace-based Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Face Recognition and Retrieval in Video

Shan¹

2010

Video Search and Mining

View full text Add to dashboard Cite

Abstract. Automatic face recognition has long been established as one of the most active research areas in computer vision. Face recognition in unconstrained environments remains challenging for most practical applications. In contrast to traditional still-image based approaches, recently the research focus has shifted towards videobased approaches. Video data provides rich and redundant information, which can be exploited to resolve the inherent ambiguities of image-based recognition like sensitivity to low resolution, pose variations and occlusion, leading to more accurate and robust recognition. Face recognition has also been considered in the content-based video retrieval setup, for example, character-based video search. In this chapter, we review existing research on face recognition and retrieval in video. The relevant techniques are comprehensively surveyed and discussed.

show abstract

“…Of the methods that compare a set to a set, canonical correlations 1 of linear subspaces have attracted much attention with the benefits of robust and computationally efficient matching when dealing with changing conditions of data acquisition and large volumes of data as input for decision making [4,5,7,8]. The previous method called Constrained Mutual Subspace Method (CMSM) finds the constrained subspace where the entire class populations exhibit small variance [8,10]. Then, each class subspace is projected on this constrained subspace to create a model and compared by canonical correlations with the new data.…”

Section: Introductionmentioning

confidence: 99%

Incremental Learning of Locally Orthogonal Subspaces for Set-based Object Recognition

Kim

Kittler²,

Cipolla³

2006

Procedings of the British Machine Vision Conference 2006

View full text Add to dashboard Cite

Orthogonal subspaces are effective models to represent object image sets (generally any high-dimensional vector sets). Canonical correlation analysis of the orthogonal subspaces provides a good solution to discriminate objects with sets of images. In such a recognition task involving image sets, an efficient learning over a large volume of image sets, which may be increasing over time, is important. In this paper, an incremental learning method of orthogonal subspaces is proposed by updating the principal components of the class correlation and total correlation matrices separately, yielding the same solution as the batch computation with far lower computational cost. A novel concept of local orthogonality is further proposed to cope with non-linear manifolds of data vectors and find a more optimal solution of orthogonal subspaces for a certain neighbouring object image sets. In the experiments using 700 face image sets, the locally orthogonal subspaces outperformed the orthogonal subspaces as well as relevant state-of-the-art methods in accuracy. Note that the locally orthogonal subspaces are also amenable to incremental updating due to their linear property.

show abstract

Face Recognition with the Multiple Constrained Mutual Subspace Method

Cited by 54 publications

References 7 publications

[Invited Paper] Recent Advances in Biometric Recognition

[Invited Paper] Recent Advances in Biometric Recognition

Face Recognition and Retrieval in Video

Incremental Learning of Locally Orthogonal Subspaces for Set-based Object Recognition

Contact Info

Product

Resources

About