A5009955709- Biometric Identification and Security
- User Authentication and Security Systems
- Face recognition and analysis
- Handwritten Text Recognition Techniques
- Hand Gesture Recognition Systems
- Natural Language Processing Techniques
- Digital Media Forensic Detection
- Generative Adversarial Networks and Image Synthesis
- Advanced Malware Detection Techniques
- ECG Monitoring and Analysis
- Gaze Tracking and Assistive Technology
- EEG and Brain-Computer Interfaces
- Topic Modeling
- Blockchain Technology Applications and Security
- Child Development and Digital Technology
- Image Processing and 3D Reconstruction
- Gait Recognition and Analysis
- Non-Invasive Vital Sign Monitoring
- Face and Expression Recognition
- Emotion and Mood Recognition
- Nutritional Studies and Diet
- Privacy-Preserving Technologies in Data
- Infant Health and Development
- COVID-19 diagnosis using AI
- Digital Mental Health Interventions
Universidad Autónoma de Madrid
2016-2025
Advanced Neural Dynamics (United States)
2024
Darmstadt University of Applied Sciences
2024
Hospital Universitario Infanta Leonor
2022
Jamia Millia Islamia
2021
The availability of large-scale facial databases, together with the remarkable progresses deep learning technologies, in particular Generative Adversarial Networks (GANs), have led to generation extremely realistic fake content, raising obvious concerns about potential for misuse. Such fostered research on manipulation detection methods that, contrary humans, already achieved astonishing results various scenarios. In this study, we focus synthesis entire images, which is a specific type...
This work proposes a novel privacy-preserving neural network feature representation to suppress the sensitive information of learned space while maintaining utility data. The new international regulation for personal data protection forces controllers guarantee privacy and avoid discriminative hazards managing users. In our approach, discrimination are related each other. Instead existing approaches aimed directly at fairness improvement, proposed enforces selected attributes. way is not...
This article presents FRCSyn-onGoing, an ongoing challenge for face recognition where researchers can easily benchmark their systems against the state of art in open common platform using large-scale public databases and standard experimental protocols. FRCSyn-onGoing is based on Face Recognition Challenge Era Synthetic Data (FRCSyn) organized at WACV 2024. first international aiming to explore use real synthetic data independently, also fusion, order address existing limitations technology....
Systems based on deep neural networks have made a breakthrough in many different pattern recognition tasks. However, the use of these systems with traditional architectures seems not to work properly when amount training data is scarce. This case on-line signature verification task. In this paper, we propose novel writer-independent Recurrent Neural Networks (RNNs) Siamese architecture whose goal learn dissimilarity metric from pairs signatures. To best our knowledge, first time recurrent...
Background- This paper summarizes the state-of-the-art and applications based on online handwritting signals with special emphasis e-security e-health fields. Methods- In particular, we focus main achievements challenges that should be addressed by scientific community, providing a guide document for future research. Conclusions- Among all points discussed in this article, remark importance of considering security, health, metadata from joint perspective. is especially critical due to double...
Deep learning has become a breathtaking technology in the last years, overcoming traditional handcrafted approaches and even humans for many different tasks. However, some tasks, such as verification of handwritten signatures, amount publicly available data is scarce, what makes difficult to test real limits deep learning. In addition lack public data, it not easy evaluate improvements novel proposed databases experimental protocols are usually considered. The main contributions this study...
Current mobile user authentication systems based on PIN codes, fingerprint, and face recognition have several shortcomings. Such limitations been addressed in the literature by exploring feasibility of passive devices through behavioral biometrics. In this line research, work carries out a comparative analysis unimodal multimodal biometric traits acquired while subjects perform different activities phone such as typing, scrolling, drawing number, tapping screen, considering touchscreen...
This article presents SVC-onGoing1, an on-going competition for on-line signature verification where researchers can easily benchmark their systems against the state of art in open common platform using large-scale public databases, such as DeepSignDB2 and SVC2021_EvalDB3, standard experimental protocols. SVC-onGoing is based on ICDAR 2021 Competition On-Line Signature Verification (SVC 2021), which has been extended to allow participants anytime. The goal evaluate limits popular scenarios...
Media forensics has attracted a tremendous attention in the last years part due to increasing concerns around DeepFakes. Since release of initial DeepFakes databases 1st generation such as UADFV and FaceForensics++ up latest 2nd Celeb-DF DFDC, many visual improvements have been carried out, making fake videos almost indistinguishable human eye. This study provides an in-depth analysis both generations terms detection performance. Two different methods are considered our experimental...
Biometrics on mobile devices has attracted a lot of attention in recent years as it is considered user-friendly authentication method. This interest also been motivated by the success Deep Learning (DL). Architectures based Convolutional Neural Networks (CNNs) and Recurrent (RNNs) have established convenience for task, improving performance robustness comparison to traditional machine learning techniques. However, some aspects must still be revisited improved. To best our knowledge, this...
ECGs have shown unique patterns to distinguish between different subjects and present important advantages compared other biometric traits. However, the lack of public data standard experimental protocols makes evaluation comparison novel ECG methods difficult. In this study, we perform extensive analysis scenarios in recognition. We consider verification identification tasks, single- multi-session settings, multi-lead recorded with traditional user-friendly devices. also ECGXtractor, a...
Among user authentication methods, behavioural biometrics has proven to be effective against identity theft as well user-friendly and unobtrusive. One of the most popular traits in literature is keystroke dynamics due large deployment computers mobile devices our society. This paper focuses on improving biometric systems free-text scenario. scenario characterised very challenging uncontrolled text conditions, influence user's emotional physical state, in-use application. To overcome these...
Large Language Models (LLMs) such as GPT developed by OpenAI, have already shown astonishing results, introducing quick changes in our society. This has been intensified the release of ChatGPT which allows anyone to interact a simple conversationalway with LLMs, without any experience field needed. As result, rapidly applied many different tasks code- and song-writer, education, virtual assistants, etc., showing impressive results for it was not trained (zero-shot learning). The present...
Despite the widespread adoption of face recognition technology around world, and its remarkable performance on current benchmarks, there are still several challenges that must be covered in more detail. This paper offers an overview Face Recognition Challenge Era Synthetic Data (FRCSyn) organized at WACV 2024. is first international challenge aiming to explore use synthetic data address existing limitations technology. Specifically, FRCSyn targets concerns related privacy issues, demographic...
Due to the technological evolution and increasing popularity of smartphones, people can access an application using authentication based on biometric approaches from many different devices. Device interoperability is a very challenging problem for biometrics, which needs be further studied. In this paper, we focus device compensation online signature verification since trait gaining significant interest in banking commercial sector last years. The proposed approach two main stages. first one...
This paper presents the first Keystroke Biometrics Ongoing Competition (KBOC) organized to establish a reproducible baseline in person authentication using keystroke biometrics. The competition has been developed BEAT platform and includes one of largest databases publicly available based on fixed text scenario. database genuine attacker sequences from 300 users acquired four different sessions distributed month time span. correspond user's name surname, therefore, each user comprises an...
This work introduces a novel DeepFake detection framework based on physiological measurement. In particular, we consider information related to the heart rate using remote photoplethysmography (rPPG). rPPG methods analyze video sequences looking for subtle color changes in human skin, revealing presence of blood under tissues. this investigate what extent is useful videos. The proposed fake detector named DeepFakesON-Phys uses Convolutional Attention Network (CAN), which extracts spatial and...
Mobile behavioral biometrics have become a popular topic of research, reaching promising results in terms authentication, exploiting multimodal combination touchscreen and background sensor data. However, there is no way knowing whether state-of-the-art classifiers the literature can distinguish between notion user device. In this article, we present new database, BehavePassDB, structured into separate acquisition sessions tasks to mimic most common aspects mobile Human-Computer Interaction...
Face recognition systems have significantly advanced in recent years, driven by the availability of large-scale datasets. However, several issues recently came up, including privacy concerns that led to discontinuation well-established public Synthetic datasets emerged as a solution, even though current synthesis methods present other drawbacks such limited intraclass variations, lack realism, and unfair representation demographic groups. This study introduces GAN-DiffFace, novel framework...
This paper describes the design, acquisition process and baseline evaluation of new e-BioSign database, which includes dynamic signature handwriting information. Data is acquired from 5 different COTS devices: three Wacom devices (STU-500, STU-530 DTU-1031) specifically designed to capture signatures handwriting, two general purpose tablets (Samsung Galaxy Note 10.1 Samsung ATIV 7). For tablets, data collected using both pen stylus also finger in order study performance verification a mobile...