A5062944593- Video Surveillance and Tracking Methods
- Particle physics theoretical and experimental studies
- Face recognition and analysis
- Particle Detector Development and Performance
- Advanced Image and Video Retrieval Techniques
- Vehicle License Plate Recognition
- Image and Object Detection Techniques
- Human Pose and Action Recognition
- Face and Expression Recognition
- Advanced Memory and Neural Computing
- Anomaly Detection Techniques and Applications
- Hand Gesture Recognition Systems
- High-Energy Particle Collisions Research
- Neutrino Physics Research
- Handwritten Text Recognition Techniques
- Autonomous Vehicle Technology and Safety
- Astrophysics and Cosmic Phenomena
- CCD and CMOS Imaging Sensors
- Dark Matter and Cosmic Phenomena
- Particle accelerators and beam dynamics
- Emotion and Mood Recognition
- Social Robot Interaction and HRI
- Superconducting Materials and Applications
- Digital Media Forensic Detection
- Precipitation Measurement and Analysis
University of Buenos Aires
2016-2024
Joint Research Center
2024
Consejo Nacional de Investigaciones Científicas y Técnicas
2012-2024
Fundación Ciencias Exactas y Naturales
2018-2022
Hospital Fernández
2014-2022
Institute of Astronomy and Space Physics
2020-2021
Instituto de Investigaciones en Ciencias de la Salud
2018-2021
Centro Científico Tecnológico - San Juan
2016-2020
Universidad Argentina de la Empresa
2012-2018
Universidad Nacional del Centro de la Provincia de Buenos Aires
2010-2011
The fully contained events detected in the NUSEX nucleon stability experiment have been analysed to search for possible anomalies fluxes of atmospheric neutrinos. measured flux muon neutrinos is very good agreement with predictions and no anomaly has found ratio between rate electron neutrino events.
We present an algorithm for the on-board vision vehicle detection problem using a cascade of boosted classifiers. Three families features are compared: rectangular filters (Haar-like features), histograms oriented gradient (HoG), and their combination (a concatenation two preceding features). A comparative study results generative (HoG discriminative features) detectors, fusion is presented. These show that combines advantages other detectors: classifiers eliminate "easily" negative examples...
In the present paper, avoidance behavior of pedestrians was characterized by controlled experiments. Several conflict situations were studied considering different flow rates and group sizes in crossing head-on configurations. Pedestrians recorded from above, individual two-dimensional trajectories their displacement recovered after image processing. Lateral swaying amplitude step lengths measured for free pedestrians, obtaining similar values to ones reported literature. Minimum distances...
This communication deals with an Oriented-Contour Point based voting algorithm for multiclass vehicle type identification (make and model). The system obtains similar results equivalent recognition frameworks different feature selections [8]. Results also show the method to be robust partial occlusion.
Spoofing attacks carried out using artificial replicas are a severe threat for fingerprint based biometric systems and, thus, require the development of effective countermeasures.One possible protection method is to implement software modules that analyze images tell live from fake samples.Most static software-based approaches in literature on various image features, each with its own strengths, weaknesses and discriminative power.Such features can be seen as different often complementary...
In this paper, we present a text detection and localization method. Our technique is based on cascade of boosted ensemble localizer uses standard image processing techniques. We propose small set features (39 in total) capable detecting various type grey level natural scene images. Two weak learners, linear discriminant function log likelihood-ratio test under gaussian assumption, are evaluated. Single combination used to form classifiers. The proposed scheme evaluated ICDAR 2003 robust...
The identification of facial expressions with human emotions plays a key role in non-verbal communication and has applications several areas. In this work, we propose descriptor based on areas angles triangles formed by the landmarks from face images. We test these descriptors for expression recognition means two different approaches. One is dynamic approach where performed Conditional Random Field (CRF) classifier. other an adaptation k-Nearest Neighbors classifier called Citation-kNN which...
Human gestures recognition is a complex visual task where motion across time distinguishes the type of action. Automatic systems tackle this problem using machine learning architectures and training datasets. In recent years, use success robust deep techniques was compatible with availability great number these sets. This paper presents SL-Animals-DVS, an event-based action dataset captured by Dynamic Vision Sensor (DVS). The DVS records humans performing sign language various animals as...
This paper provides a comparison between two of the most used visual descriptors (image features) nowadays in field object detection. The investigated image features involved Haar filters and Histogram Oriented Gradients (HoG) applied for on road vehicle Tests are very encouraging with average detection 96% realistic on-road images.
The identification of facial expressions with human emotions plays a key role in non-verbal communication and has applications several areas. In this work, we analyze two main approaches for expression recognition. One is dynamic approach introducing new simple descriptor based on the angles formed by landmarks to capture sequence. case recognition performed Conditional Random Field (CRF) classifier. An analysis most discriminative presented. other static-based appearance method. approach,...
Face recognition approaches, especially those based on deep learning models, are becoming increasingly attractive for missing person identification, due to their effectiveness and the relative simplicity of obtaining information available comparison. However, these methods still suffer from large accuracy drops when they have tackle cross-age recognition, which is most common condition face in this specific task. To address challenges, paper we investigate contribution different generative...
This work proposes a tracking-by-detection methodology for pedestrians following using targlet framework. In this framework each pedestrian is considered as an autonomous agent, denominated targlets, and modeled with state machine. The tracking procedure initialized by people detector computed on Movement Feature Space. Detector outputs generate probabilistic fields employed the of targlet, in order to obtain their individual trajectories along sequence. then analyzed off-line to: filter...
This study aims to estimate the traffic load at street intersections obtaining circulating vehicle number through image processing and pattern recognition. The algorithm detects moving objects in a view by using level lines generates new feature space called movement (MFS). MFS primitives as segments corners match model generating hypotheses. is also grouped histogram configuration histograms of oriented (HO2 L). work uses HO2 L features validate hypotheses comparing performance different...