P315 - BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group
Auflistung P315 - BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group nach Erscheinungsdatum
1 - 10 von 33
Treffer pro Seite
Sortieroptionen
- KonferenzbeitragLearning by Environment Cluster s for Face Presentation Attack Detection(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Matsunami, Tomoaki; Uchida, Hidetsugu; Abe, Narishige; Yamada, ShigefumiFace recognition has been used widely for personal authentication. However, there is a problem that it is vulnerable to a presentation attack in which a counterfeit such as a photo is presented to a camera to impersonate another person. Although various presentation attack detection methods have been proposed, these methods have not been able to sufficiently cope with the diversity of the heterogeneous environments including presentation attack instruments (PAIs) and lighting conditions. In this paper, we propose Learning by Environment Clusters (LEC) which divides training data into some clusters of similar photographic environments and trains bona-fide and attack classification models for each cluster. Experimental results using Replay-Attack, OULU-NPU, and CelebA-Spoof show the EER of the conventional method which trains one classification model from all data was 20.0%, but LEC can achieve 13.8% EER when using binarized statistical image features (BSIFs) and support vector machine used as the classification method
- KonferenzbeitragInfluence of Test Protocols on Biometric Recognition Performance Estimation(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Eglitis, Teodors; Maiorana, Emanuele; Campisi, PatrizioThe performance of a biometric system is commonly evaluated by the obtained recognition rates and comparing the results against the ones reported in the literature on the same database. An aspect that has not received the deserved attention in the literature concerns the influence, on the achieved rates, of the test protocol employed to select the enrol and probe data. We provide a detailed analysis of the impact of the experimental choices on the estimated performance, considering the recommendations provided by ISO/IEC 19795 standard. We use the UTFVP finger vein database, reproducing results presented in the literature using multiple protocols. Our experiments highlight the possibility of obtaining equal error rates reduced by half simply by changing the test protocol.
- KonferenzbeitragOn Recognizing Occluded Faces in the Wild(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Erakın, Mustafa Ekrem; Demir, Uğur; Ekenel, Hazım KemalFacial appearance variations due to occlusion has been one of the main challenges for face recognition systems. To facilitate further research in this area, it is necessary and important to have occluded face datasets collected from real-world, as synthetically generated occluded faces cannot represent the nature of the problem. In this paper, we present the Real World Occluded Faces (ROF) dataset, that contains faces with both upper face occlusion, due to sunglasses, and lower face occlusion, due to masks. We propose two evaluation protocols for this dataset. Benchmark experiments on the dataset have shown that no matter how powerful the deep face representation models are, their performance degrades significantly when they are tested on real-world occluded faces. It is observed that the performance drop is far less when the models are tested on synthetically generated occluded faces. The ROF dataset and the associated evaluation protocols are publicly available at the following link https://github.com/ekremerakin/RealWorldOccludedFaces.
- KonferenzbeitragOn the Relevance of Minutiae Count and Distribution for Finger Vein Recognition Accuracy(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Linortner, Michael; Uhl, AndreasVein recognition usually uses binary features, but besides deep learning-based approaches key-point and minutiae-based ones started to become popular as well. Statistical measures for vein minutiae points, like spatial point distribution, have not been investigated in literature so far. In this work the number of vein minutiae points and their spatial distribution is analyzed in relation to recognition accuracy. The goal is to initiate a discussion on statistical behavior of vein minutiae points and deriving possible quality measures for vein minutiae point sets.
- KonferenzbeitragMy Eyes Are Up Here: Promoting Focus on Uncovered Regions in Masked Face Recognition(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Neto, Pedro; Boutros, Fadi; Pinto, João Ribeiro; Saffari, Mohsen; Damer, Naser; Sequeira, Ana F.; Cardoso, Jaime S.The recent Covid-19 pandemic and the fact that wearing masks in public is now mandatory in several countries, created challenges in the use of face recognition systems (FRS). In this work, we address the challenge of masked face recognition (MFR) and focus on evaluating the verification performance in FRS when verifying masked vs unmasked faces compared to verifying only unmasked faces. We propose a methodology that combines the traditional triplet loss and the mean squared error (MSE) intending to improve the robustness of an MFR system in the masked-unmasked comparison mode. The results obtained by our proposed method show improvements in a detailed step-wise ablation study. The conducted study showed significant performance gains induced by our proposed training paradigm and modified triplet loss on two evaluation databases.
- KonferenzbeitragTransferability Analysis of an Adversarial Attack on Gender Classification to Face Recognition(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Rezgui, Zohra; Bassit, AminaModern biometric systems establish their decision based on the outcome of machine learning (ML) classifiers trained to make accurate predictions. Such classifiers are vulnerable to diverse adversarial attacks, altering the classifiers' predictions by adding a crafted perturbation. According to ML literature, those attacks are transferable among models that perform the same task. However, models performing different tasks, but sharing the same input space and the same model architecture, were never included in transferability scenarios. In this paper, we analyze this phenomenon for the special case of VGG16-based biometric classifiers. Concretely, we study the effect of the white-box FGSM attack, on a gender classifier and compare several defense methods as countermeasure. Then, in a black-box manner, we attack a pre-trained face recognition classifier using adversarial images generated by the FGSM. Our experiments show that this attack is transferable from a gender classifier to a face recognition classifier where both were independently trained.
- KonferenzbeitragImage Quality Assessment on Identity Documents(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Yáñez, Claudio; Tapia, JuanThis paper developed a method for performing Image Quality Assessment (IQA) on ID-Card images. First, we build the dataset, consisting of 204 images from Chilean ID-Cards, containing real and tampered images with varying quality levels. Then, we evaluated different features, obtaining the best results using the BRISQUE features and a newly trained SVR, with an $R^2$ score of 0.5868. This proposed method is called BRISQUE-ID. The IQA on ID-Cards can be used as a pre-processing stage for discarding lousy quality images and helping the subsequent steps in the processing pipeline.
- KonferenzbeitragOn Brightness Agnostic Adversarial Examples Against Face Recognition Systems(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Singh, Inderjeet; Momiyama, Satoru; Kakizaki, Kazuya; Araki, ToshinoriThis paper introduces a novel adversarial example generation method against face recognition systems (FRSs). An adversarial example (AX) is an image with deliberately crafted noise to cause incorrect predictions by a target system. The AXs generated from our method remain robust under real-world brightness changes. Our method performs non-linear brightness transformations while leveraging the concept of curriculum learning during the attack generation procedure. We demonstrate that our method outperforms conventional techniques from comprehensive experimental investigations in the digital and physical world. Furthermore, this method enables practical risk assessment of FRSs against brightness agnostic AXs.
- KonferenzbeitragEvaluation on Biometric Accuracy Estimation Using Generalized Pareto (GP) Distribution(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Yamada, Shigefumi; Matsunami, TomoakiThe accuracy of biometric authentication technology is becoming more sophisticated with its progress. For this reason, a huge number of biometric samples are required for accuracy evaluation, and the increased collection cost is an issue for biometric vendors. This work establishes a biometric accuracy estimation method using an extreme value theory to reduce the collection cost. It also explains the estimation procedure of false match rate using the generalized Pareto distribution and shows results applied to the face, gait, and voice comparison score data with an estimation effect of about 5–10 times. We investigate the criteria for the applicability of extremum statistics through application cases.
- KonferenzbeitragInteroperability of Contact and Contactless Fingerprints Across Multiple Fingerprint Sensors(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Williams, Brady; McCauley, John; Dando, John; Nasrabadi, Nasser; Dawson, JeremyContactless fingerprinting devices have grown in popularity in recent years due to speed and convenience of capture. Also, due to the global COVID-19 pandemic, the need for safe and hygienic options for fingerprint capture are more pressing than ever. However, contactless systems face challenges in the areas of interoperability and matching performance as shown in other works. In this paper, we present a contactless vs. contact interoperability assessment of several contactless devices, including cellphone fingerphoto capture. In addition to evaluating the match performance of each contactless sensor, this paper presents an analysis of the impact of finger size and skin melanin content on contactless match performance. AUC results indicate that contactless match performance of the newest contactless devices is reaching that of contact fingerprints. In addition, match scores indicate that, while not as sensitive to melanin content, contactless fingerprint matching may be impacted by finger size.