A Classification-Selection Approach for Self Updating of Face Verification Systems Under Stringent Storage and Computational Requirements

Tuveri, Pierluigi; Mura, Valerio; Marcialis, Gian Luca; Roli, Fabio

doi:10.1007/978-3-319-23234-8_50

Cited by 5 publications

(7 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First of all, we implemented standard facial recognition as a sort of "ground truth", since our main claim is to clarify to what extent self update approaches are helpful over top-performing deep-learning based face recognition. In particular, we tested the "traditional" self-update system [30], the classification-selection method based on risk minimization [31] and two methods of classification-selection with limited number of templates per user, based on K-means [32] and on the semi-supervised application of RANDOM editing methods just to verify if the selection performed by the two methods above is significant or not. In fact, it is possible to categorize biometric adaptive systems into two categories:…”

Section: Adaptive Methods For Template Updatementioning

confidence: 99%

“…In particular, Rattani et al introduced a classification/selection system based on harmonic functions and a risk minimization technique [31]. In [32] the authors present a method that keeps the number of templates constant at each iteration by setting the maximum number of images per user, namely, p, in the selection phase. This is obtained by deriving the centroid of the samples of a given subject through the K-means algorithm, and then selecting the p closest samples to that centroid.…”

Section: Adaptive Methods For Template Updatementioning

confidence: 99%

See 1 more Smart Citation

Are Adaptive Face Recognition Systems still Necessary? Experiments on the APE Dataset

Orrù,

Micheletto,

Fierrez

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

In the last five years, deep learning methods, in particular CNN, have attracted considerable attention in the field of face-based recognition, achieving impressive results. Despite this progress, it is not yet clear precisely to what extent deep features are able to follow all the intra-class variations that the face can present over time. In this paper we investigate the performance the performance improvement of face recognition systems by adopting self updating strategies of the face templates. For that purpose, we evaluate the performance of a well-known deep-learning face representation, namely, FaceNet, on a dataset that we generated explicitly conceived to embed intra-class variations of users on a large time span of captures: the APhotoEveryday (APE) dataset 1 . Moreover, we compare these deep features with handcrafted features extracted using the BSIF algorithm. In both cases, we evaluate various template update strategies, in order to detect the most useful for such kind of features. Experimental results show the effectiveness of "optimized" self-update methods with respect to systems without update or random selection of templates.

show abstract

Section: Adaptive Methods For Template Updatementioning

confidence: 99%

Section: Adaptive Methods For Template Updatementioning

confidence: 99%

Are Adaptive Face Recognition Systems still Necessary? Experiments on the APE Dataset

Orrù,

Micheletto,

Fierrez

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We extract four frames from each video, and rotate and scale face images to have eyes in the same positions. We use the same matcher described in [42]. It accounts for illumination variations as in [43], and then computes a BSIF descriptor [44].…”

Section: Methodsmentioning

confidence: 99%

Statistical Meta-Analysis of Presentation Attacks for Secure Multibiometric Systems

Biggio

Fumera

Marcialis

et al. 2017

IEEE Trans. Pattern Anal. Mach. Intell.

Self Cite

View full text Add to dashboard Cite

Abstract-Prior work has shown that multibiometric systems are vulnerable to presentation attacks, assuming that their matching score distribution is identical to that of genuine users, without fabricating any fake trait. We have recently shown that this assumption is not representative of current fingerprint and face presentation attacks, leading one to overestimate the vulnerability of multibiometric systems, and to design less effective fusion rules. In this paper, we overcome these limitations by proposing a statistical meta-model of face and fingerprint presentation attacks that characterizes a wider family of fake score distributions, including distributions of known and, potentially, unknown attacks. This allows us to perform a thorough security evaluation of multibiometric systems against presentation attacks, quantifying how their vulnerability may vary also under attacks that are different from those considered during design, through an uncertainty analysis. We empirically show that our approach can reliably predict the performance of multibiometric systems even under never-before-seen face and fingerprint presentation attacks, and that the secure fusion rules designed using our approach can exhibit an improved trade-off between the performance in the absence and in the presence of attack. We finally argue that our method can be extended to other biometrics besides faces and fingerprints.

show abstract

“…These reasons explain why, to the best of our knowledge, no self-update algorithms have been implemented and integrated yet in real face verification applications. Therefore, in this paper, as follow-up of [13], we faced with the self-update problem by considering a very small number of templates per client. Besides the advantage of meeting eventual and stringent hardware requirements of mobile devices, the limited number of templates may also reduce the probability of introducing impostors into the users' gallery.…”

Section: Introductionmentioning

confidence: 99%

“…With regard to our previous work [13], we updated and increased the evidence reported in that early publication by: (1) clearly explaining the rationale behind our approach, that was only drafted in that early publication; (2) proposing two more algorithms; (3) performing a large set of experiments which simulate conditions near and far from the system's working hypothesis. This allowed to clarify when the proposed methods can work and when not; (4) comparing our algorithms' performance with that of other existing self-updating approaches [9,12,14], in order to assess their possible advantages and drawbacks with respect to the state of the art.…”

Section: Introductionmentioning

confidence: 99%

A novel classification-selection approach for the self updating of template-based face recognition systems

Orrù

Marcialis

Roli

2020

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

The boosting on the need of security notably increased the amount of possible facial recognition applications, especially due to the success of the Internet of Things (IoT) paradigm. However, although handcrafted and deep learning-inspired facial features reached a significant level of compactness and expressive power, the facial recognition performance still suffers from intra-class variations such as ageing, facial expressions, lighting changes, and pose. These variations cannot be captured in a single acquisition and require multiple acquisitions of long duration, which are expensive and need a high level of collaboration from the users. Among others, self-update algorithms have been proposed in order to mitigate these problems. Self-updating aims to add novel templates to the users' gallery among the inputs submitted during system operations. Consequently, computational complexity and storage space tend to be among the critical requirements of these algorithms. The present paper deals with the above problems by a novel template-based self-update algorithm, able to keep over time the expressive power of a limited set of templates stored in the system database. The rationale behind the proposed approach is in the working hypothesis that a dominating mode characterises the features' distribution given the client. Therefore, the key point is to select the best templates around that mode. We propose two methods, which are tested on systems based on handcrafted features and deep-learning-inspired autoencoders at the state-of-the-art. Three benchmark data sets are used. Experimental results confirm that, by effective and compact feature sets which can support our working hypothesis, the proposed classification-selection approaches overcome the problem of manual updating and, in case, stringent computational requirements.

show abstract

A Classification-Selection Approach for Self Updating of Face Verification Systems Under Stringent Storage and Computational Requirements

Cited by 5 publications

References 5 publications

Are Adaptive Face Recognition Systems still Necessary? Experiments on the APE Dataset

Are Adaptive Face Recognition Systems still Necessary? Experiments on the APE Dataset

Statistical Meta-Analysis of Presentation Attacks for Secure Multibiometric Systems

A novel classification-selection approach for the self updating of template-based face recognition systems

Contact Info

Product

Resources

About