P339 - BIOSIG 2023 - Proceedings of the 22nd International Conference of the Biometrics Special Interest Group
Auflistung nach:
Auflistung P339 - BIOSIG 2023 - Proceedings of the 22nd International Conference of the Biometrics Special Interest Group nach Erscheinungsdatum
1 - 10 von 33
Treffer pro Seite
Sortieroptionen
- KonferenzbeitragAssessing the Human Ability to Recognize Synthetic Speech in Ordinary Conversation(BIOSIG 2023, 2023) Daniel Prudký, Anton FircThis work assesses the human ability to recognize synthetic speech (deepfake). This paper describes an experiment in which we communicated with respondents using voice messages. We presented the respondents with a cover story about testing the user-friendliness of voice messages while secretly sending them a pre-prepared deepfake recording during the conversation. We examined their reactions, knowledge of deepfakes, or how many could correctly identify which message was deepfake. The results show that none of the respondents reacted in any way to the fraudulent deepfake message, and only one retrospectively admitted to noticing something specific. On the other hand, a voicemail message that contained a deepfake was correctly identified by 83.9% of respondents after revealing the nature of the experiment. Thus, the results show that although the deepfake recording was clearly identifiable among others, no one reacted to it. In summary, we show that the human ability to recognize voice deepfakes is not at a level we can trust. It is very difficult for people to distinguish between real and fake voices, especially if they do not expect them.
- KonferenzbeitragOn the Impact of Tattoos on Hand Recognition(BIOSIG 2023, 2023) Lazaro Janier Gonzalez-Soler, Kacper Marek ZylaFrom Native Americans, who used tattoos as a way of seducing the opposite sex, to prisoners in the last century, who were identified by tattooed numbers, tattoos have been used for many years for a variety of purposes. Nowadays, tattoos express affiliation or beliefs and can therefore serve as complementary information to identify individuals. To support forensic investigations, hand-based biometrics have emerged as a promising technology to recognise individuals. As several statistics have reported an increase in the use of tattoos on hands, in this paper, we investigate the impact of tattoos on the performance of state-of-the-art hand recognition systems. To this end, we first propose a method for generating semi-synthetic tattooed hands. A benchmark is then performed for tattooed and non-tattooed hands. Experimental results computed on a freely available database showed that, although in some cases the use of tattoos assists hand recognition, the observed trend is a deterioration of recognition accuracy, indicating the sensitivity of hand recognition systems to tattoos.
- KonferenzbeitragFace verification explainability heatmap generation using a vision transformer(BIOSIG 2023, 2023) Ricardo Correia, Paulo L CorreiaExplainable Face Recognition (XFR) is a critical technology to support the large deployment of learning-based face recognition solutions. This paper aims at contributing to the more transparent usage of Vision Transformers (ViTs) for face verification (FV) tasks, by proposing a novel approach for generating FV explainability heatmaps, for both positive and negative decisions. The proposed solution leverages on the attention maps generated by a ViT and employs masking techniques to create masks based on the highlighted regions in the attention maps. These masks are applied to the pair of faces, and the masking technique with most impact on the decision is selected to be used to generate heatmaps for the probe-gallery pair of faces. These heatmaps offer valuable insights into the decision-making process, shedding light on the most important face regions for the verification outcome. The key novelty of this paper lies in the proposed approach for generating explainability heatmaps tailored for verification pairs in the context of ViT models, which combines the ViT attention maps regions of the probe-gallery pair to create masks that allow evaluating those region´s impact on the verification decision for both positive and negative decisions.
- KonferenzbeitragUnified Face Image Quality Score based on ISO/IEC Quality Components(BIOSIG 2023, 2023) Praveen Kumar Chandaliya, Kiran RajaFace image quality assessment is crucial in the face enrolment process to obtain high-quality face images in the reference database. Neglecting quality control will adversely impact the accuracy and efficiency of face recognition systems, resulting in an image captured with poor perceptual quality. In this work, we present a holistic combination of $21$ component quality measures proposed in ``ISO/IEC CD 29794-5" and identify the varying nature of different measures across different datasets. The variance is seen across both capture-related and subject-related measures, which can be tedious for validating each component metric by a human observer when judging the quality of the enrolment image. Motivated by this observation, we propose an efficient method of combining quality components into one unified score using a simple supervised learning approach. The proposed approach for predicting face recognition performance based on the obtained unified face image quality assessment (FIQA) score was comprehensively evaluated using three datasets representing diverse quality factors. We extensively evaluate the proposed approach using the Error-vs-Discard Characteristic (EDC) and show its applicability using five different FRS. The evaluation indicates promising results of the proposed approach combining multiple component scores into a unified score for broader application in face image enrolment in FRS.
- KonferenzbeitragRobust Sclera Segmentation for Skin-tone Agnostic Face Image Quality Assessment(BIOSIG 2023, 2023) Wassim Kabbani, Christoph BuschFace image quality assessment (FIQA) is crucial for obtaining good face recognition performance. FIQA algorithms should be robust and insensitive to demographic factors. The eye sclera has a consistent whitish color in all humans regardless of their age, ethnicity and skin-tone. This work proposes a robust sclera segmentation method that is suitable for face images in the enrolment and the border control face recognition scenarios. It shows how the statistical analysis of the sclera pixels produces features that are invariant to skin-tone, age and ethnicity and thus can be incorporated into FIQA algorithms to make them agnostic to demographic factors.
- KonferenzbeitragLVT Face Database: A benchmark database for visible and hidden face biometrics(BIOSIG 2023, 2023) Nélida Mirabet-Herranz, Jean-Luc DugelayAlthough the estimation of eHealth parameters from face visuals (images and videos) has grown as a major area of research in the past years, deep-learning-based models are still challenged by RGB lack of robustness, for instance with changing illumination conditions. As a means to overcome these limitations and to unlock new opportunities, thermal imagery has arisen as a favorable alternative to solidify different technologies such as heart rate estimation from faces. However, the reduced number of databases containing thermal imagery and the lack of health annotation of the subjects in them limits the exploration of this spectrum. Motivated by this, in this paper, we present our Label-EURECOM Visible and Thermal (LVT) Face Database for face biometrics. This database is the first that contains paired visible and thermal images and videos from 52 subjects with metadata of 22 soft biometrics and health parameters. Moreover, we establish the first study introducing the potential of thermal images for weight estimation from faces on our database.
- KonferenzbeitragHuman-centered evaluation of anomalous events detection in crowded environments(BIOSIG 2023, 2023) Giulia Orrù, Elia PorceddaAnomaly detection in crowd analysis refers to the ability to detect events and people’s behaviours that deviate from normality. Anomaly detection techniques are developed to support human operators in various monitoring and investigation activities. So far, the anomaly detectors' performance evaluation derives from the rate of correctly classified individual frames, according to the labels given by the annotator. This evaluation does not make the system's performance appreciable, especially from a human operator viewpoint. In this paper, we propose a novel evaluation approach called ``Trigger-Level evaluation'' that is shown to be human-centered and closer to the user's perception of the system's performance. In particular, we define two new performance metrics to aid the evaluation of the usability of anomaly detectors in real-time.
- KonferenzbeitragCyclist Recognition from a Silhouette Set(BIOSIG 2023, 2023) Eijiro Makishima, Fumito ShinmuraPerson recognition from surveillance cameras can be useful for criminal investigations. Currently, gait recognition technology can identify walking individuals, but recognition of people riding bicycles has not been actively investigated, despite cycling being a popular mode of transportation. In this paper, we propose a method to recognize individuals riding bicycles (cyclists) using a silhouette set. We captured two types of cyclist data, normal and rush modes, from five different views, and generated silhouette image sequences from this data. We evaluated accuracy of the proposed method on the silhouette images in identification and verification tasks. The evaluation results demonstrate the effectiveness of our proposed method.
- KonferenzbeitragComparison of two architectures for text-independent verification after character-unaware text segmentation(BIOSIG 2023, 2023) Maria De Marsico, Mohammadreza ShabaniThis paper compares the performance of two popular CNN architectures, ResNet-50 and MobileNetV2, fine-tuned for text-independent writer verification. The used benchmark is IAM dataset. The further contributions are an easy and fast sub-region cropping for robust model training, and a biometrics-oriented performance evaluation. The preliminary results are encouraging.
- KonferenzbeitragFace Image De-identification Based on Feature Embedding for Privacy Protection(BIOSIG 2023, 2023) Goki Hanawa, Koichi ItoWith the expansion of social networking services, a large number of face images have been disclosed on the Internet.Since face recognition makes it easy to collect face images of specific persons, the collected face images can be used to attack face recognition systems, such as spoofing attacks.Face image de-identification, which makes face recognition difficult without changing the appearance of the face image, is necessary for disclosing face images safely on the Internet.In this paper, we propose a face image de-identification method by embedding facial features of another person into a face image.The proposed method uses a convolutional neural network to generate a face image that can be recognized as that of another person while preserving the appearance of the face image.Through a set of experiments using a public face image dataset, we demonstrate that the proposed method preserves the appearance of face images and has high de-identification performance against unknown face recognition models compared to conventional methods.