A5013928164- Face recognition and analysis
- Face and Expression Recognition
- Biometric Identification and Security
- Image Retrieval and Classification Techniques
- Digital Media Forensic Detection
- Emotion and Mood Recognition
- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Human Pose and Action Recognition
- Generative Adversarial Networks and Image Synthesis
- Anomaly Detection Techniques and Applications
- User Authentication and Security Systems
- Gait Recognition and Analysis
- Speech and Audio Processing
- COVID-19 diagnosis using AI
- Time Series Analysis and Forecasting
- Advanced Image Processing Techniques
- Remote-Sensing Image Classification
- Hand Gesture Recognition Systems
- Infant Health and Development
- Stock Market Forecasting Methods
- Traffic Prediction and Management Techniques
- Gaze Tracking and Assistive Technology
- Medical Image Segmentation Techniques
- Music and Audio Processing
Sorbonne University Abu Dhabi
2022-2025
Sorbonne Université
2022-2024
University of Oulu
2013-2022
École Centrale de Lille
2019-2022
Institut d'électronique de microélectronique et de nanotechnologie
2019-2022
Université de Lille
2019-2022
Centre National de la Recherche Scientifique
2019-2022
Université Polytechnique Hauts-de-France
2019-2022
Northwestern Polytechnical University
2016-2019
Centre de Développement des Technologies Avancées
2000
This paper presents a novel and efficient facial image representation based on local binary pattern (LBP) texture features. The face is divided into several regions from which the LBP feature distributions are extracted concatenated an enhanced vector to be used as descriptor. performance of proposed method assessed in recognition problem under different challenges. Other applications extensions also discussed.
Current face biometric systems are vulnerable to spoo ing attacks. A spoofing attack occurs when a person tries masquerade as someone else by falsifying data and thereby gaining illegitimate access. Inspired image quality assessment, characterization of printing artifacts, differences in light reflection, we propose approach the problem detection from texture analysis point view. Indeed, prints usually contain defects that can be well detected using features. Hence, present novel based on...
Research on non-intrusive software-based face spoofing detection schemes has been mainly focused the analysis of luminance information images, hence discarding chroma component, which can be very useful for discriminating fake faces from genuine ones. This paper introduces a novel and appealing approach detecting using colour texture analysis. We exploit joint colour-texture chrominance channels by extracting complementary low-level feature descriptions different spaces. More specifically,...
The vulnerabilities of face-based biometric systems to presentation attacks have been finally recognized but yet we lack generalized software-based face attack detection (PAD) methods performing robustly in practical mobile authentication scenarios. This is mainly due the fact that existing public PAD datasets are beginning cover a variety scenarios and acquisition conditions their standard evaluation protocols do not encourage researchers assess generalization capabilities across these...
Research on face spoofing detection has mainly been focused analyzing the luminance of images, hence discarding chrominance information which can be useful for discriminating fake faces from genuine ones. In this work, we propose a new anti-spoofing method based color texture analysis. We analyze joint color-texture and channels using local binary pattern descriptor. More specifically, feature histograms are extracted each image band separately. Extensive experiments two benchmark datasets,...
Abstract Human activity recognition (HAR) systems attempt to automatically identify and analyze human activities using acquired information from various types of sensors. Although several extensive review papers have already been published in the general HAR topics, growing technologies field as well multi-disciplinary nature prompt need for constant updates field. In this respect, paper attempts summarize progress computer vision perspective. Indeed, most applications such interaction,...
The face recognition community has finally started paying more attention to the long-neglected problem of spoofing attacks and number countermeasures is gradually increasing. Fairly good results have been reported on publicly available databases but it reasonable assume that there exists no superior anti-spoofing technique due varying nature attack scenarios acquisition conditions. Therefore, we propose approach as a set attack-specific subproblems are solvable with proper combination...
Recently deep Convolutional Neural Networks have been successfully applied in many computer vision tasks and achieved promising results. So some works introduced the learning into face anti-spoofing. However, most approaches just use final fully-connected layer to distinguish real fake faces. Inspired by idea of each convolutional kernel can be regarded as a part filter, we extract partial features from neural network (CNN) In our prosed approach, CNN is fine-tuned firstly on spoofing...
The vulnerabilities of face biometric authentication systems to spoofing attacks have received a significant attention during the recent years. Some proposed countermeasures achieved impressive results when evaluated on intratests, i.e., system is trained and tested same database. Unfortunately, most these techniques fail generalize well unseen attacks, e.g., one database then another This major concern in antispoofing research that mostly overlooked. In this letter, we propose novel...
Abstract User authentication is an important step to protect information, and in this context, face biometrics potentially advantageous. Face natural, intuitive, easy use, less human-invasive. Unfortunately, recent work has revealed that vulnerable spoofing attacks using cheap low-tech equipment. This paper introduces a novel appealing approach detect the spatiotemporal (dynamic texture) extensions of highly popular local binary pattern operator. The key idea learn structure dynamics facial...
This paper presents a novel fully automatic bi-modal, face and speaker, recognition system which runs in real-time on mobile phone. The implemented Nokia N900 demonstrates the feasibility of performing both speaker We evaluate this publicly-available phone database provide well defined evaluation protocol. was captured almost exclusively using phones aims to improve research into deploying biometric techniques devices. show, database, that can be performed environment score fusion...
Biometrics already form a significant component of current and emerging identification technologies. systems aim to determine or verify the identity an individual from their behavioral and/or biological characteristics. Despite progress, some biometric fail meet multitude stringent security robustness requirements support deployment in practical scenarios. Among concerns are vulnerabilities spoofing?persons who masquerade as others gain illegitimate accesses protected data, services,...
We introduce a novel discriminative feature space which is efficient not only for face detection but also recognition. The representation based on local binary patterns (LBP) and consists of encoding both global facial characteristics into compact histogram. proposed invariant with respect to monotonic gray scale transformations can be derived in single scan through the image. Considering space, second-degree polynomial kernel SVM classifier was trained detect frontal faces images....
Current face biometric systems are vulnerable to spoofing attacks. A attack occurs when a person tries masquerade as someone else by falsifying data and thereby gaining illegitimate access. Inspired image quality assessment, characterisation of printing artefacts differences in light reflection, the authors propose approach problem detection from texture analysis point view. Indeed, prints usually contain defects that can be well detected using local shape features. Hence, present novel...
Automatic medical diagnosis is an emerging center of interest in computer vision as it provides unobtrusive objective information on a patient's condition. The face, mirror health status, can reveal symptomatic indications specific diseases. Thus, the detection facial abnormalities or atypical features at upmost importance when comes to diagnostics. This survey aims give overview recent developments diagnostics from images based methods. Various approaches have been considered assess...
In recent years, software-based face presentation attack detection (PAD) methods have seen a great progress. However, most existing schemes are not able to generalize well in more realistic conditions. The objective of this competition is evaluate and compare the generalization performances mobile PAD techniques under some real-world variations, including unseen input sensors, instruments (PAI) illumination conditions, on larger scale OULU-NPU dataset using its standard evaluation protocols...
Recent experiments, reported in the third edition of Fingerprint Liveness Detection competition (LivDet 2013), have clearly shown that fingerprint liveness detection is a very difficult and challenging task. Although number approaches large, none them can be claimed as able to detect traits with an acceptable error rate. In our opinion, order investigate at which extent this reduced, novel feature sets must proposed, and, eventually, integrated existing ones. paper, descriptor named "BSIF"...
Spoofing identities using photographs is one of the most common techniques to attack 2-D face recognition systems. There seems exist no comparative studies different same protocols and data. The motivation behind this competition compare performance state-of-the-art algorithms on database a unique evaluation method. Six teams from universities around world have participated in contest. Use or multiple motion, texture analysis liveness detection appears be trend competition. Most are able...
Facial occlusions, due for example to sunglasses, hats, scarf, beards etc., can significantly affect the performance of any face recognition system. Unfortunately, presence facial occlusions is quite common in real-world applications especially when individuals are not cooperative with system such as video surveillance scenarios. While there has been an enormous amount research on under pose/illumination changes and image degradations, problems caused by mostly overlooked. The focus this...
The face recognition community has finally started paying more attention to the long-neglected problem of spoofing attacks. number countermeasures is gradually increasing and fairly good results have been reported on publicly available databases. There exists no superior antispoofing technique due varying nature attack scenarios acquisition conditions. Therefore, it important find out complementary study how they should be combined in order construct an easily extensible anti-spoofing...