Face template protection using deep LDPC codes learning

Chen, Lingying; Zhao, Guanghui; Zhou, Junwei; Ho, Anthony T. S.; Cheng, L.

doi:10.1049/iet-bmt.2018.5156

Cited by 28 publications

(18 citation statements)

References 57 publications

(85 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The match rate refers to the proportion of P s for which the comparison score between V * and V was greater than or equal to a pre-defined match threshold. The match rate was computed at two thresholds established on the baseline systems' development set 14 of face embeddings: at FMR = 0.1% (commonly used) and at FMR = 1%, to represent higher-security and lower-security application scenarios, respectively.…”

Section: Tablementioning

confidence: 99%

Towards Protecting Face Embeddings in Mobile Face Verification Scenarios

Hahn,

Marcel

2021

Preprint

View full text Add to dashboard Cite

This paper proposes PolyProtect, a method for protecting the sensitive face embeddings that are used to represent people's faces in neural-network-based face verification systems. PolyProtect transforms a face embedding to a more secure template, using a mapping based on multivariate polynomials parameterised by user-specific coefficients and exponents. In this work, PolyProtect is evaluated on two open-source face verification systems in a mobile application context, under the toughest threat model that assumes a fully-informed attacker with complete knowledge of the system and all its parameters. Results indicate that PolyProtect can be tuned to achieve a satisfactory trade-off between the recognition accuracy of the PolyProtected face verification system and the irreversibility of the PolyProtected templates. Furthermore, PolyProtected templates are shown to be effectively unlinkable, especially if the user-specific parameters employed in the PolyProtect mapping are selected in a non-naive manner. The evaluation is conducted using practical methodologies with tangible results, to present realistic insight into the method's robustness as a face embedding protection scheme in practice. The code to fully reproduce this work is available at: https://gitlab.idiap.ch/bob/bob.paper.polyprotect 2021.

show abstract

Section: Tablementioning

confidence: 99%

Towards Protecting Face Embeddings in Mobile Face Verification Scenarios

Hahn,

Marcel

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Examples of NN-learned face BTP methods that pre-define a representative binary code (and cryptographically hash it to produce the protected template), include [37]- [42]. Of these, [37]- [39] adopt an image-level approach, whereby the NN is trained in an end-to-end manner to map the input face images to their corresponding, pre-defined codes. Alternatively, [40]- [42] adopt a feature-level approach, where the NN training starts from the extracted face features as opposed to the raw images.…”

Section: Feature-levelmentioning

confidence: 99%

“…The binary codes obtained during enrollment and authentication are cryptographically hashed, and comparison is based on an exact match between the two hashes. A similar approach is adopted in [39], except that the user-specific binary codes are encoded into error-correcting codes (LDPC) during training, such that the CNN learns the mapping between a user's face image and their corresponding LDPC code. During authentication, the LDPC code obtained for the probe image is decoded to recover the underlying binary code, and the cryptographic hashes of the probe and reference binary codes are compared.…”

Section: Feature-levelmentioning

confidence: 99%

“…Table II [37]- [39] [43], [44], [50], [51] [45]- [49] C. Non-NN versus NN-learned Face BTP Methods Research into both method types increased from 2018 to 2019, with a slight preference for Non-NN BTP methods. In 2020, there seems to have been a noticeable increase in the preference for Non-NN over NN-learned face BTP methods, probably due to the availability of several pre-trained DNN models that could be used as pre-BTP feature extractors.…”

Section: Feature-levelmentioning

confidence: 99%

See 1 more Smart Citation

Biometric Template Protection for Neural-Network-based Face Recognition Systems: A Survey of Methods and Evaluation Techniques

Hahn¹,

Marcel²

2021

Preprint

View full text Add to dashboard Cite

As automated face recognition applications tend towards ubiquity, there is a growing need to secure the sensitive face data used within these systems. This paper presents a survey of biometric template protection (BTP) methods proposed for securing face "templates" (images or representative features) in neural-network-based face recognition systems. The BTP methods are categorised into two types: Non-NN and NN-learned. Non-NN methods use a neural network (NN) as a feature extractor, but the BTP part is based on a non-NN algorithm applied at either image-level or feature-level. In contrast, NN-learned methods specifically employ a NN to learn a protected template from the unprotected face image or features. We present examples of Non-NN and NN-learned face BTP methods from the literature, along with a discussion of the two categories' comparative strengths and weaknesses. We also investigate the techniques used to evaluate these BTP methods, in terms of the three most common BTP criteria: "recognition accuracy", "irreversibility", and "renewability/unlinkability". As expected, the recognition accuracy of protected face recognition systems is generally evaluated using the same (empirical) techniques employed for evaluating standard (unprotected) biometric systems. On the contrary, most irreversibility and renewability/unlinkability evaluations are found to be based on theoretical assumptions/estimates or verbal implications, with a lack of empirical validation in a practical face recognition context. We recommend, therefore, a greater focus on empirical evaluation strategies, to provide more concrete insights into the irreversibility and renewability/unlinkability of face BTP methods in practice. Additionally, an exploration of the reproducibility of the studied BTP works, in terms of the public availability of their implementation code and evaluation datasets/procedures, suggests that it would currently be difficult for the BTP community to faithfully replicate (and thus validate) most of the reported findings. So, we advocate for a push towards reproducibility, in the hope of furthering our understanding of the face BTP research field.

show abstract

“…Chen et al [9] suggested a comprehensive face template protection scheme to secure the original face template. The facial feature of each is mapped to a different binary code in the training using deep multilabel learning.…”

Section: Review Criteriamentioning

confidence: 99%

Comparison of levels and fusion approaches for multimodal biometrics

S¹,

Reddy²

2021

IJEECS

View full text Add to dashboard Cite

The biometric-based authentication system occupies maximal space in the field of security administration. Biometric applications are swiftly accelerating in day-to-day life such as computer login, smart homes, online banking, hospitals, border areas, industries, forensics, e-voting attendance system and investigation of crime. A reliable and accurate recognition body can be achieved with multimodal biometric methodologies. In this paper, we discuss starting with an introduction to biometric systems followed by their classification, and advantages as well as disadvantages. In today’s world, most of the systems are unimodal biometrics having a lot of limitations to overcome those multimodal biometrics comes in to picture. In this paper we have discussed comprehensive representation on the system of multimodal biometric, various modes of undertakings, the significance of information fusion, a different section is allotted on the various possible levels of fusion involving sensor-level, feature-level, score-level, and decision -level as well as different rules of fusion.

show abstract

Face template protection using deep LDPC codes learning

Cited by 28 publications

References 57 publications

Towards Protecting Face Embeddings in Mobile Face Verification Scenarios

Towards Protecting Face Embeddings in Mobile Face Verification Scenarios

Biometric Template Protection for Neural-Network-based Face Recognition Systems: A Survey of Methods and Evaluation Techniques

Comparison of levels and fusion approaches for multimodal biometrics

Contact Info

Product

Resources

About