Multi-class Fukunaga Koontz discriminant analysis for enhanced face recognition

Juefei-Xu, Felix; Savvides, Marios

doi:10.1016/j.patcog.2015.10.007

Cited by 30 publications

(8 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also a few studies [32,112,113] utilizing Nearest Neighbor models on this topic. Júnior et al [20] pro-posed the Nearest Neighbor Distance Ratio (NNDR) classifier, which in turn, is a multiclass open-set extension for the Nearest Neighbor (NN) algorithm and is referred to as Open Set NN (OSNN).…”

Section: Statistical Approachesmentioning

confidence: 99%

“…An automatic face recognition system that is usually encountered with unknown individuals [111,113,[163][164][165][166][167][168] is another domain to be deployed in open-universe scenarios. There is a wide range of real applications of face recognition, for instance, reducing retail crime, controlling mobile phone access, helping police officers, identifying people on social media platforms, and so on.…”

Section: Applications Of Open Set Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey on Open Set Recognition

Mahdavi,

Carvalho

2021

Preprint

View full text Add to dashboard Cite

Recognition (OSR) is about dealing with unknown situations that were not learned by the models during training. In this paper, we provide a survey of existing works about OSR and distinguish their respective advantages and disadvantages to help out new researchers interested in the subject. The categorization of OSR models is provided along with an extensive summary of recent progress. Additionally, the relationships between OSR and its related tasks including multi-class classification and novelty detection are analyzed. It is concluded that OSR can appropriately deal with unknown instances in the real-world where capturing all possible classes in the training data is not practical. Lastly, applications of OSR are highlighted and some new directions for future research topics are suggested.

show abstract

Section: Statistical Approachesmentioning

confidence: 99%

Section: Applications Of Open Set Recognitionmentioning

confidence: 99%

A Survey on Open Set Recognition

Mahdavi,

Carvalho

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The reconstructed image patchŷ = Dx. The aforementioned dictionary learning and reconstruction were previously used in various domain-domain mapping problems such as [3,[22][23][24][25][26][27][28].…”

Section: Shallow Reconstruction From Dense Vs Sparse Representationmentioning

confidence: 99%

FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction

Huang

Juefei-Xu

Wang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

Self Cite

View full text Add to dashboard Cite

At this moment, GAN-based image generation methods are still imperfect, whose upsampling design has limitations in leaving some certain artifact patterns in the synthesized image. Such artifact patterns can be easily exploited (by recent methods) for difference detection of real and GAN-synthesized images. However, the existing detection methods put much emphasis on the artifact patterns, which can become futile if such artifact patterns were reduced. Towards reducing the artifacts in the synthesized images, in this paper, we devise a simple yet powerful approach termed FakePolisher that performs shallow reconstruction of fake images through a learned linear dictionary, intending to effectively and efficiently reduce the artifacts introduced during image synthesis. In particular, we first train a dictionary model to capture the patterns of real images. Based on this dictionary, we seek the representation of DeepFake images in a low dimensional subspace through linear projection or sparse coding. Then, we are able to perform shallow reconstruction of the 'fake-free' version of the DeepFake image, which largely reduces the artifact patterns DeepFake introduces. The comprehensive evaluation on 3 state-of-the-art DeepFake detection methods and fake images generated by 16 popular GAN-based fake image generation techniques, demonstrates the effectiveness of our technique. Overall, through reducing artifact patterns, our technique significantly reduces the accuracy of the 3 state-of-theart fake image detection methods, i.e., 47% on average and up to 93% in the worst case. Our results confirm the limitation of current fake detection methods and calls the attention of DeepFake researchers and practitioners for more general-purpose fake detection techniques. CCS CONCEPTS • Security and privacy → Human and societal aspects of security and privacy; • Computing methodologies → Computer vision.

show abstract

“…Face and iris modalities include a very interesting area of biometric systems [29][30][31][32][33][34][35][36]. All face images in this study are resized to 60 × 60 undergo histogram equalisation and mean-variance normalisation [37].…”

Section: Unimodal Biometric Systemsmentioning

confidence: 99%

Optimum scheme selection for face–iris biometric

Eskandari

Sharifi

2017

IET biom.

View full text Add to dashboard Cite

Designing a new dynamic and optimal scheme for face-iris fusion based on the score level, feature level and decision level fusion is considered in this study. Prior to implementing the proposed combined level fusion, several schemes are separately implemented at each level of fusion to investigate the performance improvement of each level of fusion on face and iris modalities. In fact, the optimum scheme is constructed by selecting flexible and dynamic features and scores of face and iris biometrics and then combining the advantages of different levels of fusion. Consequently, the scheme produces a set of fast and flexible features and scores for fusion. On the other hand, the idea of threshold-optimised decisions is used in this study to fuse the optimised decisions of face and iris biometrics. Experimental results on verification rates demonstrate a significant improvement of proposed combined level fusion scheme over unimodal and multimodal fusion methods.

show abstract

Multi-class Fukunaga Koontz discriminant analysis for enhanced face recognition

Cited by 30 publications

References 59 publications

A Survey on Open Set Recognition

A Survey on Open Set Recognition

FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction

Optimum scheme selection for face–iris biometric

Contact Info

Product

Resources

About