Emanuel Aldea

Email : emanuel.$lastname$ @ u-psud.fr
Address : Bureau 2069, Lab. SATIE, Batiment 660 DIGITEO, rue Noetzlin, Gif-sur-Yvette, 91190 France
Directions : if you visit our offices, check the guide

I am working as an Associate Professor at Paris Sud University, in the MOSS (Methods and Systems for Signal Processing) group of SATIE. I completed my PhD in 2009 at Télécom ParisTech, under the supervision of Isabelle Bloch.
My work draws from the areas of computer vision and machine learning, applied to a variety of problems in image understanding, cartography and 3D reconstruction.
Trying to schedule a meeting with me? Check my availability here!

Some news

04.2020
Gianni will present TRADI at the GdR-ISIS meeting "Apprentissage profond et modèles génératifs pour modéliser l'incertitude des données"
12.2019
The project MUSical Instrument Conservation with Optical Monitoring (MUSICOM) submitted in collaboration with our colleagues from Laboratorio Arvedi di Diagnostica Non-Invasiva presso il Museo del Violino has been funded by the French-Italian PHC Galilée partnership
01.2019
Submissions are open for a special issue of Sensors on deep learning for multi-sensor fusion. Submission deadline: July 1, 2019
05.2019
Congratulations to Jennifer Vandoni for defending her PhD, and all the best for the future!
04.2019
Pint of Science France organizes during the 20-22 May Festival a scientific popularization evening related to the study of crowds, where I will talk about observing crowds. More details on the registration page here
01.2019
We welcome in our group Gianni Franchi, who will be involved as postdoc in the MOHICANS project



Current research activities

A significant part of my research time is devoted to these topics, and I will be happy to further discuss any of them if your research interests are potentially related. If you are a student, please read the available information in the Students section first.


High-density crowd analysis cams
Keywords: multiple camera networks, wide baseline calibration, segmentation, tracking, fluid dynamics

The figure on the right shows the crowd in the holy Muslim site of Makkah in October 2012 during the peak of the pilgrimage, filmed with our camera system. The broad purpose of this research project is to enable us to understand in detail the mechanisms and the evolution of human interactions in high-density crowds (more than 4-5 people per square meter). The multiple camera setup allows us to overcome the issues created by occlusion and to have extra information for performing reliable tracking in these difficult conditions. There are interesting projects focusing on modeling animal behaviour in large, dense groups; similarly, once highly accurate individual tracking is performed in dense human crowds, we expect to obtain surprising results about the apparition of crowd instability that will be very helpful in preventing loss of human life, and in infrastructure planning.
Starting 11.2015, the ANR project MOHICANS supports this research. More details are provided here.


Real-time cartography and pose estimation dashcam2 dashcam2
Keywords: robotics, SLAM, bundle adjustment, sensor fusion

I am interested in working on navigation problems related to image processing and data fusion, and in finding solutions that may leverage some limitations of existing SLAM algorithms. The applications in this domain are various: autonomous robots and UAVs, intelligent vehicles, cartography, pose estimation, servoing, planning etc. I am particularly interested in modelling how different sensors exhibit varying performances depending on complex factors such as ego-motion, scene etc. An example of this kind of work is SuperFAST.


NDE for infrastructure assessment detection crack
Keywords: crack detection, non destructive examination, image processing, a-contrario, marked point processes, UAV

Our society will benefit significantly from the large scale automated NDE which is foreseeable in the light of the current advances in robotic vision. I am interested in improving the performance of currently available algorithms for the detection of cracks in the specific context of aerial inspection, which is often characterized by image quality degradation. Our team evaluated different families of, and proposed, algorithms based on minimal cost path analysis, on image percolation, on marked point processes and on a-contrario strategies. In the publications, we highlight their respective advantages and limitations (with respect to detection accuracy, degradations such as motion and focus blur, computational cost and potential for real time processing etc).

Previous research activities

Monocular depth field estimation with known camera motion scene
Keywords: depth maps, optical flow, asymptotic observers, TV-L1 regularization

The objective of this work is to reconstruct the dense structure of a static scene observed by a monocular camera system following a known trajectory. One of our contributions is represented by providing a TV-L1 energy functional that estimates directly the unknown depth field given the camera motion, thus avoiding to estimate as an intermediate step an optical flow field with additional geometric constraints. Our method has two main interests: we highlight a practical minimal parametrization for the given assumptions (static scene, known camera motion) and we solve the resulting variational problem using an efficient, discontinuity preserving formulation. In this work, we also propose a solution for integrating the sequence of depth observations using asymptotic observers. The scheme is derived from a suitable optical flow formulation that highlights efficiently the known camera motion. The motion parameters may be obtained from inertial sensors, from a motion capture system or if applicable from a visual odometry algorithm.


Multi-modal tracking lips
Keywords: tracking, particle filters, multiple appearance models

In this work, we propose a novel method to track an object whose appearance is evolving in time. The tracking procedure is performed by a particle filter algorithm in which all possible appearance models are explicitly considered using a mixture decomposition of the likelihood. Then, the component weights of this mixture are conditioned by both the state and the current observation. Moreover, the use of the current observation makes the estimation process more robust and allows handling complementary features, such as color and shape information. In the proposed approach, these estimated component weights are computed using a Support Vector Machine (but unsupervised learning might be a very interesting possibility too). Tests on a mouth tracking problem show that the multiple appearance model outperforms classical single appearance likelihood.




Structured data learning for image interpretation fuzzy
Keywords: structured data learning, graphical models, graph kernels, support vector machines, image interpretation, multiple kernel learning, fuzzy logic, spatial relations

Image interpretation methods use primarily the visual features of low-level or high-level interest elements. However, spatial information concerning the relative positioning of these elements is equally beneficial, as it has been shown previously in segmentation and structure recognition. Therefore, an interest for the integration of spatial information in the learning framework has emerged recently. The fact that spatial information is often perceived and expressed in a manner which is close to natural language, along with the fact that the absence of a spatial interaction is also relevant, hint at the usefulness of fuzzy spatial information for image representation. Fuzzy representations actually permit to assess at the same time the imprecision degree of a relation (e.g., ``close to'' or ``to the left of'') and the gradual transition between the satisfiability and the non-satisfiability of a relation. Among the solutions used to adapt image data to algorithm inputs, we adopt a representation structure which encodes explicitly image parts and spatial interactions in a graphical model.
The objective of this work is to explore techniques of spatial information representation and their integration in the learning process, within the context of image classifiers that make use of graph kernels. We motivate our choice of labeled graphs for representing images, in th e context of learning with SVM classifiers. Graph kernels have been studied intensively in computational chemistry and the study of biologic systems, and an adaptation for image related graphs is necessary, since image structures and properties of the information encoded in the labeling are fundamentally different. We illustrate the integration of spatial information within the image graphical model by considering fuzzy adjacency measures between interest elements (regions), and we define a family of graph representations determined by different thresholds applied to these spatial measures. Finally, we employ multiple kernel learning methods in order to build up classifiers that can take into account different graphical representations of the same image at once. The results show that spatial information complements the visual features of distinctive elements in images and that adapting the discriminative kernel functions for the fuzzy spatial representations is beneficial in terms of performance.
Available positions

For any Master internship position available, the information will be posted here. Please do not contact me for undergraduate level internships, and/or shorter than 5-6 months (the only exception: you already interacted with me personally, or a professor I know may provide a recommendation for you). I do my best to provide some feedback to prospective interns, but if your profile/application is pretty much unrelated to the internship description, my reply will have a low priority.
Available Master internship positions :
  • none for the moment


Available PhD positions :
  • none for the moment


Available postdoctoral position :
  • none for the moment


Current and past students

PhD Students :
  • Alireza Rezaei, with Sylvie Le Hégarat-Mascle, Historical music instrument 3D reconstruction and wear detection in UV-induced fluorescence photography , Oct. 2018 -
  • Huiqin Chen, with Sylvie Le Hégarat-Mascle, Egocentric video registration for collaborative localization , Dec. 2017 -
  • Jennifer Vandoni, with Sylvie Le Hégarat-Mascle, Ensemble Methods for Pedestrian Detection in Dense Crowds , Dec. 2015 - May 2019
  • Nicola Pellicanò, with Sylvie Le Hégarat-Mascle, Tackling pedestrian detection in large scenes with multiple views and representations , Oct. 2015 - Dec. 2018

Postdoc Researchers :
  • Gianni Franchi, Large scale tracking , Jan. 2019 - Feb. 2020

Master internships :
  • Xuanlong Yu (Télécom Saint Etienne, M2) Enhanced annotation of pedestrian data in video sequences witth active learning , March-August 2020
  • Zhuzhi Fan (INSA Lyon - Paris Sud University, M2) Localization of a mobile camera wearer for security in urban environments , March-August 2019
  • Corentin Presvots (ENS Paris Saclay, M1), with LMS, Ecole Polytechnique A vision-based study of auxetic meta-materials , May-July 2018
  • Yixuan Yao (Paris Sud University, M1) Flying object detection for locating mobile cameras , June-July 2018
  • Kevin Mercier (Paris Sud University, M1), with Guillaume Charpiat Deep network optimization for crowd analysis , May-Aug. 2018
  • Thi-Hao Nguyen (Japan Advanced Institute of Science and Technology, M2), with LMS, Ecole Polytechnique 3D registration for the deformation of nanomaterials , March-Aug. 2018
  • Alireza Rezaei (University Jean Monnet, M2) GPU acceleration and 3D rendering for multiple camera pedestrian detection , Jan.-July 2018
  • Raphaël Guegan (Paris Sud University, M1) with Guillaume Charpiat, Réseaux de neurones pour l’étude de la dynamique des foules , May-Sept. 2017
  • Camille Palmier (Paris Sud University, M1) with Paola Goatin, Validation de modèles macroscopiques non-locaux de dynamique de foule , May-Sept. 2017
  • Lucile Denet (Télécom Physique Strasbourg, M2) with Thales Avionics, Hybridation inertie et vision pour des applications aéronautiques , Mar.-Aug. 2016
  • Sen Yan (Paris Sud University, M1), Inference de la structure de la scene pour l'analyse multi-camera d'espaces urbains , May.-Jul. 2016
  • Huiqin Chen (Paris Sud University, M1), Analyse du flot des foules denses pour la caractérisation de leur dynamique , Jun.-Jul. 2015
  • Victor Truong (ENS Cachan, M1), High speed IMU-camera registration , May-Jul. 2015
  • El Mehdi Abdali (Paris Sud University, M2) with Theraclion, HIFU propagation analysis for non-invasive tumor therapy , Mar.-Aug. 2015
  • Davide Marastoni (University of Pavia, M2), High-density Crowd Segmentation , Mar.-Jul. 2014
  • Romain Saussard (INSA Lyon, M2), Cartographie étendue par fusion de cartes locales à l'aide de capteurs inertiels , Feb.-Aug. 2012

Short Master projects :
  • Zhuzhi Fan, Mohammed Chghaf (Paris Sud University, M2), Calibration of a wide baseline multiple camera system , 2019
  • Hacène Karrad, Yassine Ouali, Jingkun Yang (Paris Sud University, M2) with Jean-Pierre Barbot and Isabelle Vin, Indoor geolocalization with wi-fi based fingerprinting and vision , 2018
  • Christophe Ribal, Victor Truong (ENS Cachan, M2), Reduction de dimmensionalité pour la détection de têtes , 2017
  • Wei Liu, Le Sun (Paris Sud University, M2), La détection de fissures en utilisant l'algorithme FOSA , 2016
  • Kenza Benseghir, Zineb Saadaoui (Paris Sud University, M2), Implémentation de l'algorithme EDLines , 2016
  • Chun Geng, Xiao-Kang Ma (Paris Sud University, M1), Surveillance d’abeilles en vol à l’aide de la vision par ordinateur , 2016
  • Wen Dai, Sen Yan (Paris Sud University, M1), Étude de la fréquence du battement des ailes d’une abeille avec des caméras à fréquence variable , 2016
  • Yali Zhu, Wei Liu (Paris Sud University, M1), L'algorithme de percolation pour la detection de fissures en NDE , 2015
  • Hongyi Wang, Yameng Li (Paris Sud University, M1), Suivi autonome d'une cible par un quadri-rotor , 2014
  • Siyuan Liu (Paris Sud University, M1), Etude d'images de hybridation fluorescente in situ (FISH) pour la détection d'anomalies génétiques , 2014
  • Mehdi Sadouni (Paris 6 University, M2), Apprentissage pour la détection de points d'intérêt et descripteurs locaux robustes en temps réel , 2011
  • Romain Bachy (Paris 6 University, M2), Recherche d’images similaires à partir de descripteurs couleurs, 2010


Teaching

I am currently involved in the following courses (academic year 2019-2020):
M1 E3A Orsay Cachan
443 - Object oriented programming and algorithm design (CM : 21h, TP : 24h, TD : 5h)
M1 E3A Orsay Cachan
453 - Image and signal processing, together with Thomas Rodet (CM : 20h, TP : 16h, TD : 14h)
M1 E3A Orsay Cachan
471 - Study and research project (TP : 10-20h)
M2 E3A SETI
501 - C++ intensive course (TP : 14h)
M2 E3A SETI
554 - Robotic vision, together with Sylvie Le Hégarat-Mascle (CM : 18h, TP : 6h)
M2 E3A SETI
519 - Research seminar on ethics in robotics research, together with Alain Mérigot (CM : 3.5h)
M1 ERGO
107 - Photometry and colorimetry (CM : 24.5h)
DU OC2
Photometry and colorimetry (CM : 24.5h)
Polytech Paris Sud
5APP Image processing (CM : 8h, TP : 8h, TD : 8h)
Pavia Univ., CS Master
Computer vision (CM : 12h)
Publications

TRADI: Tracking deep neural network weight distributions, Gianni Franchi, Andrei Bursuc, Emanuel Aldea, Séverine Dubuisson and Isabelle Bloch, preprint, 2019 online version, slides, bib
@article{franchi19arxiv,
  title={TRADI: Tracking deep neural network weight distributions},
  author={Franchi, Gianni and Bursuc, Andrei and Aldea, Emanuel and Dubuisson, S{\'e}verine and Bloch, Isabelle},
  journal={arXiv preprint arXiv:1912.11316},
  year={2019}
}
Use of Scene Geometry Priors for Data Association in Egocentric Views, Huiqin Chen, Emanuel Aldea, Sylvie Le Hégarat-Mascle and Vincent Despiegel, Proceedings of the 8th International Workshop on Biometrics and Forensics (IWBF2020), 2020 draft, bib
@inproceedings{chen20iwbf,
 author = {Chen, Huiqin and Aldea, Emanuel and Le H{\'e}garat-Mascle, Sylvie and Despiegel, Vincent },
 booktitle = {Proceedings of the 8th International Workshop on Biometrics and Forensics (IWBF2020)},
 title = {Use of Scene Geometry Priors for Data Association in Egocentric Views},
 year = {2020}
}
Augmenting Deep Learning Performance in an Evidential Multiple Classifier System , Jennifer Vandoni, Sylvie Le Hgarat-Mascle and Emanuel Aldea, Sensors , 2019 online version, bib
@article{vandoni19sensors,
AUTHOR = {Vandoni, Jennifer and H{\'e}garat-Mascle, Sylvie Le and Aldea, Emanuel},
TITLE = {Augmenting Deep Learning Performance in an Evidential Multiple Classifier System},
JOURNAL = {Sensors},
VOLUME = {19},
YEAR = {2019},
NUMBER = {21},
ARTICLE-NUMBER = {4664},
URL = {https://www.mdpi.com/1424-8220/19/21/4664},
ISSN = {1424-8220},
DOI = {10.3390/s19214664}
}

Constraining Relative Camera Pose Estimation with Pedestrian Detector-Based Correspondence Filters, Emanuel Aldea, Thomas Pollok and Chengchao Qu, Proceedings of the 16th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS), 2019 draft, bib
@inproceedings{aldea19avss, 
  author={E. {Aldea} and T. {Pollok} and C. {Qu}}, 
  booktitle={2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)}, 
  title={Constraining Relative Camera Pose Estimation with Pedestrian Detector-Based Correspondence Filters}, 
  year={2019}, 
  volume={}, 
  number={}, 
  pages={1-7}
} 
Crowd Behaviour Characterization for Scene Tracking , Gianni Franchi, Emanuel Aldea, Séverine Dubuisson and Isabelle Bloch, Proceedings of the 16th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS), 2019 draft, bib
@inproceedings{franchi19avss, 
  author={G. {Franchi} and E. {Aldea} and S. {Dubuisson} and I. {Bloch}}, 
  booktitle={2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)}, 
  title={Crowd Behavior Characterization for Scene Tracking}, 
  year={2019}, 
  volume={}, 
  number={}, 
  pages={1-8}
} 
Determining Epipole Location Integrity by Multimodal Sampling, Huiqin Chen, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the 16th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS), The 3th International Workshop on Traffic and Street Surveillance for Safety and Security (IWT4S), 2019 draft, bib
@inproceedings{chen19avss, author={H. {Chen} and E. {Aldea} and S. L. {Hégarat-Mascle}}, 
  booktitle={2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)}, 
  title={Determining Epipole Location Integrity by Multimodal Sampling}, 
  year={2019}, 
  volume={}, 
  number={}, 
  pages={1-8}
} 
Detecting alterations in historical violins with optical monitoring, Alireza Rezaei, Emanuel Aldea, Piercarlo Dondi, Marco Malagodi and Sylvie Le Hégarat-Mascle, Proceedings of the 14th International Conference on Quality Control by Artificial Vision (QCAV), 2019 draft, bib
@inproceedings{rezaei19qcav,
 author = {Rezaei, Alireza and Aldea, Emanuel and Dondi, Piercarlo and Malagodi, Marco and Le H{\'e}garat-Mascle, Sylvie   },
 booktitle = {Proceedings of the 14th International Conference on Quality Control by Artificial Vision (QCAV)},
 year = {2019},
 doi = {10.1117/12.2521702},
 URL = {https://doi.org/10.1117/12.2521702},
 title = {Detecting alterations in historical violins with optical monitoring}
}
Efficient evaluation of the Number of False Alarm criterion , Sylvie Le Hgarat-Mascle , Emanuel Aldea and Jennifer Vandoni EURASIP Journal on Image and Video Processing , 2019 online version, bib
@article{hegarat19jivp,
  author    = {Sylvie Le H{\'{e}}garat{-}Mascle and
               Emanuel Aldea and
               Jennifer Vandoni},
  title     = {Efficient evaluation of the Number of False Alarm criterion},
  journal   = {{EURASIP} J. Image and Video Processing},
  volume    = {2019},
  pages     = {35},
  year      = {2019},
  url       = {https://doi.org/10.1186/s13640-019-0429-4}
}

Integrating Visual and Geometric Consistency for Pose Estimation, Huiqin Chen, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the 16th International Conference on Machine Vision Applications (MVA), 2019 draft, bib
@inproceedings{chen19mva,
 author={H. {Chen} and E. {Aldea} and S. {Le Hégarat-Mascle}},
 booktitle={2019 16th International Conference on Machine Vision Applications (MVA)},
 title={Integrating Visual and Geometric Consistency for Pose Estimation},
 year={2019},
 pages={1-5},
 doi={10.23919/MVA.2019.8757911},
 ISSN={},
 month={May}
}
Wide baseline pose estimation from video with a density-based uncertainty model, Nicola Pellicanò, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Machine Vision and Applications, 2019 draft, full-text online, bib
@article{Pellicano2019,
author="Pellican{\`o}, Nicola
and Aldea, Emanuel
and Le H{\'e}garat-Mascle, Sylvie",
title="Wide baseline pose estimation from video with a density-based uncertainty model",
journal="Machine Vision and Applications",
year="2019",
month="Jun",
day="13",
abstract="Robust wide baseline pose estimation is an essential step in the deployment of smart camera networks. In this work, we highlight some current limitations of conventional strategies for relative pose estimation in difficult urban scenes. Then, we propose a solution which relies on an adaptive search of corresponding interest points in synchronized video streams which allows us to converge robustly toward a high-quality solution. The core idea of our algorithm is to build across the image space a nonstationary mapping of the local pose estimation uncertainty, based on the spatial distribution of interest points. Subsequently, the mapping guides the selection of new observations from the video stream in order to prioritize the coverage of areas of high uncertainty. With an additional step in the initial stage, the proposed algorithm may also be used for refining an existing pose estimation based on the video data; this mode allows for performing a data-driven self-calibration task for stereo rigs for which accuracy is critical, such as onboard medical or vehicular systems. We validate our method on three different datasets which cover typical scenarios in pose estimation. The results show a fast and robust convergence of the solution, with a significant improvement, compared to single image-based alternatives, of the RMSE of ground-truth matches, and of the maximum absolute error.",
issn="1432-1769",
doi="10.1007/s00138-019-01036-6",
url="https://doi.org/10.1007/s00138-019-01036-6"
}
Evaluating Crowd Density Estimators via Their Uncertainty Bounds , Jennifer Vandoni, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the 26th International Conference on Image Processing (ICIP) , 2019 draft, bib
@inproceedings{vandoni19icip,
 author = {Vandoni, Jennifer and Aldea, Emanuel and Le H{\'e}garat-Mascle, Sylvie   },
 booktitle = {Proceedings of the 26th International Conference on Image Processing (ICIP) },
 title = {Evaluating Crowd Density Estimators via Their Uncertainty Bounds},
 year = {2019}
}
Evidential Query-By-Committee Active Learning for Pedestrian Detection in High-Density Crowds, Jennifer Vandoni, Emanuel Aldea and Sylvie Le Hégarat-Mascle, International Journal of Approximate Reasoning, 2019 draft, bib
@article{vandoni19ijar,
 title = "Evidential query-by-committee active learning for pedestrian detection in high-density crowds",
journal = "International Journal of Approximate Reasoning",
volume = "104",
pages = "166 - 184",
year = "2019",
issn = "0888-613X",
doi = "https://doi.org/10.1016/j.ijar.2018.11.007",
url = "http://www.sciencedirect.com/science/article/pii/S0888613X18303517",
author = "Jennifer Vandoni and Emanuel Aldea and Sylvie Le Hégarat-Mascle"
}
2CoBel: A Scalable Belief Function Representation for 2D Discernment Frames, Nicola Pellicanò, Sylvie Le Hégarat-Mascle and Emanuel Aldea, International Journal of Approximate Reasoning, 2018 draft, bib
@article{pellicano18ijar,
 author = {Pellican{\`o}, Nicola and Le H{\'e}garat-Mascle, Sylvie  and Aldea, Emanuel },
 title = "2CoBel: A scalable belief function representation for 2D discernment frames",
 journal = "International Journal of Approximate Reasoning",
 volume = "103",
 pages = "320 - 342",
 year = "2018",
 issn = "0888-613X",
 doi = "https://doi.org/10.1016/j.ijar.2018.10.007",
 url = "http://www.sciencedirect.com/science/article/pii/S0888613X18303864"
}
GPU-accelerated Height Map Estimation with Local Geometry Priors in Large Scenes, Alireza Rezaei, Nicola Pellicanò, and Emanuel Aldea, Proceedings of the 15th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS), 2018 draft, bib
@inproceedings{rezaei18avss,
 author = {Rezaei, Alireza and Pellican{\`o}, Nicola and Aldea, Emanuel },
 booktitle = {Proceedings of the 15th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS)},
 title = {GPU-accelerated Height Map Estimation with Local Geometry Priors in Large Scenes},
 year = {2018}
}
Evidential Split and Merge: Application to Object-Based Image Analysis, Marie Lachaize, Sylvie Le Hégarat-Mascle, Emanuel Aldea, Aude Maitrot and Roger Reynaud, International Journal of Approximate Reasoning, draft, bib
@article{lachaize18ijar,
 title = "Evidential split-and-merge: Application to object-based image analysis",
 author = "Marie Lachaize and Sylvie Le H{\'e}garat-Mascle and Emanuel Aldea and Aude Maitrot and Roger Reynaud",
 journal = "International Journal of Approximate Reasoning",
 volume = "103",
 pages = "303 - 319",
 year = "2018",
 issn = "0888-613X",
 doi = "https://doi.org/10.1016/j.ijar.2018.10.008",
 url = "http://www.sciencedirect.com/science/article/pii/S0888613X18302986"
}
2CoBel: An Efficient Belief Function Extension for Two-dimensional Continuous Spaces, Nicola Pellicanò, Sylvie Le Hégarat-Mascle and Emanuel Aldea, Proceedings of the 21st International Conference on Information Fusion (FUSION), 2018 draft, bib
@inproceedings{pellicano18fusion,
 author = {Pellican{\`o}, Nicola and Le H{\'e}garat-Mascle, Sylvie  and Aldea, Emanuel },
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (FUSION)},
 title = {2{C}o{B}el: An Efficient Belief Function Extension for Two-dimensional Continuous Spaces},
 pages={1032-1039},
 doi={10.23919/ICIF.2018.8455783}, 
 year = {2018}
}
Belief Function Definition for Ensemble Methods - Application to Pedestrian Detection in Dense Crowds, Jennifer Vandoni, Sylvie Le Hégarat-Mascle and Emanuel Aldea, Proceedings of the 21st International Conference on Information Fusion (FUSION), 2018 draft, bib
@inproceedings{vandoni18fusion,
 author = {Vandoni, Jennifer and Le H{\'e}garat-Mascle, Sylvie  and Aldea, Emanuel },
 booktitle = {Proceedings of the 21st International Conference on Information Fusion (FUSION)},
 title = {Belief Function Definition for Ensemble Methods - Application to Pedestrian Detection in Dense Crowds},
 pages={2481-2488}, 
 doi={10.23919/ICIF.2018.8455313}, 
 year = {2018}
}
Evidential framework for Error Correcting Output Code classification, Marie Lachaize, Sylvie Le Hégarat-Mascle, Emanuel Aldea, Aude Maitrot and Roger Reynaud, Engineering Applications of Artificial Intelligence , 2018 final draft, bib
@article{lachaize18eaai,
 title = "Evidential framework for {E}rror {C}orrecting {O}utput {C}ode classification",
journal = "Engineering Applications of Artificial Intelligence",
volume = "73",
pages = "10 - 21",
year = "2018",
issn = "0952-1976",
doi = "https://doi.org/10.1016/j.engappai.2018.04.019",
url = "http://www.sciencedirect.com/science/article/pii/S0952197618300988",
author = "Marie Lachaize and Sylvie Le H{\'e}garat-Mascle and Emanuel Aldea and Aude Maitrot and Roger Reynaud"
}
A novel approach for multi-object tracking using evidential representation for objects, Wafa Rekik, Sylvie Le Hégarat-Mascle and Emanuel Aldea, Proceedings of the 20th International Conference on Information Fusion (FUSION), 2017 final draft, bib
@inproceedings{rekik17fusion,
 author    = {Wafa Rekik and
               Sylvie Le H{\'{e}}garat{-}Mascle and
               Emanuel Aldea},
  title     = {A novel approach for multi-object tracking using evidential representation
               for objects},
  booktitle = {20th International Conference on Information Fusion, {FUSION} 2017,
               Xi'an, China, July 10-13, 2017},
  pages     = {1--8},
  year      = {2017},
  url       = {https://doi.org/10.23919/ICIF.2017.8009819},
  doi       = {10.23919/ICIF.2017.8009819}
}
An Evidential Framework for Pedestrian Detection in High-Density Crowds, Jennifer Vandoni, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the 14th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS), 2017 final draft, bib
@inproceedings{vandoni17avss,
 author    = {Jennifer Vandoni and
               Emanuel Aldea and
               Sylvie Le H{\'{e}}garat{-}Mascle},
  title     = {An evidential framework for pedestrian detection in high-density crowds},
  booktitle = {14th {IEEE} International Conference on Advanced Video and Signal
               Based Surveillance, {AVSS} 2017, Lecce, Italy, August 29 - September
               1, 2017},
  pages     = {1--6},
  year      = {2017},
  url       = {https://doi.org/10.1109/AVSS.2017.8078498},
  doi       = {10.1109/AVSS.2017.8078498}
}
Active Learning for High-Density Crowd Count Regression, Jennifer Vandoni, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the 14th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS) - 2nd Workshop on Signal Processing for Understanding Crowd Dynamics, 2017 final draft, bib
@inproceedings{vandoni17spcrowd,
 author    = {Jennifer Vandoni and
               Emanuel Aldea and
               Sylvie Le H{\'{e}}garat{-}Mascle},
  title     = {Active learning for high-density crowd count regression},
  booktitle = {14th {IEEE} International Conference on Advanced Video and Signal
               Based Surveillance, {AVSS} 2017, Lecce, Italy, August 29 - September
               1, 2017},
  pages     = {1--6},
  year      = {2017},
  url       = {https://doi.org/10.1109/AVSS.2017.8078508},
  doi       = {10.1109/AVSS.2017.8078508}
}
Geometry-Based Multiple Camera Head Detection in Dense Crowds, Nicola Pellicanò, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the 28th British Machine Vision Conference (BMVC) - 5th Activity Monitoring by Multiple Distributed Sensing Workshop, 2017 final draft, bib
@inproceedings{pellicano17ammds,
 author = {Pellican{\`o}, Nicola and Aldea, Emanuel and Le H{\'e}garat-Mascle, Sylvie},
 booktitle = {Proceedings of the 28th British Machine Vision Conference (BMVC) - 5th Activity Monitoring by Multiple Distributed Sensing Workshop},
 title = {Geometry-Based Multiple Camera Head Detection in Dense Crowds},
 year = {2017}
}
Pressure Estimation In A High-Density Crowd Using A Multi-Scale Count Regressor, Emanuel Aldea and Khurom H. Kiyani, Proceedings of the 12th International Conference on Traffic and Granular Flow (TGF), 2017 bib
@inproceedings{aldea17tgf,
 author = {Aldea, Emanuel and Kiyani, Khurom H.},
 booktitle = {Proceedings of the 12th International Conference on Traffic and Granular Flow (TGF)},
 title = {Pressure Estimation In A High-Density Crowd
Using A Multi-Scale Count Regressor},
 year = {2017}
}
Evidential multi-class classification from binary classifiers: application to waste sorting quality control from hyperspectral data, Marie Lachaize, Sylvie Le Hégarat-Mascle, Emanuel Aldea, Aude Maitrot and Roger Reynaud, Proceedings of the 13th International Conference on Quality Control by Artificial Vision (QCAV), 2017 final draft, bib
@inproceedings{lachaize17qcav,
 author    = {Marie Lachaize and
               Sylvie Le H{\'{e}}garat{-}Mascle and
               Emanuel Aldea and
               Aude Maitrot and
               Roger Reynaud},
  title     = {Evidential multi-class classification from binary classifiers: application
               to waste sorting quality control from hyperspectral data},
  booktitle = {Thirteenth International Conference on Quality Control by Artificial
               Vision, {QCAV} 2017, Tokyo, Japan, May 14, 2017},
  pages     = {103380V},
  year      = {2017},
  url       = {https://doi.org/10.1117/12.2266961},
  doi       = {10.1117/12.2266961}
}
Crack Detection Based on a Marked Point Process Model, Jennifer Vandoni, Sylvie Le Hégarat-Mascle and Emanuel Aldea, Proceedings of the International Conference on Pattern Recognition (ICPR), 2016 final draft, supp. material, bib
@inproceedings{vandoni16icpr,
 author    = {Jennifer Vandoni and
               Sylvie Le H{\'{e}}garat{-}Mascle and
               Emanuel Aldea},
  title     = {Crack detection based on a {M}arked {P}oint {P}rocess model},
  booktitle = {23rd International Conference on Pattern Recognition, {ICPR} 2016,
               Canc{\'{u}}n, Mexico, December 4-8, 2016},
  pages     = {3933--3938},
  year      = {2016}
}
Robust Wide Baseline Pose Estimation from Video, Nicola Pellicanò, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the International Conference on Pattern Recognition (ICPR), 2016 final draft, supp. material, bib
@inproceedings{pellicano16icpr,
 author    = {Nicola Pellican{\`o} and
               Emanuel Aldea and
               Sylvie Le H{\'{e}}garat{-}Mascle},
  title     = {Robust wide baseline pose estimation from video},
  booktitle = {23rd International Conference on Pattern Recognition, {ICPR} 2016,
               Canc{\'{u}}n, Mexico, December 4-8, 2016},
  pages     = {3820--3825},
  year      = {2016}
}
HOOFR: An Enhanced Bio-Inspired Feature Extractor, Dai-Duong Nguyen, Abdelhafid El Ouardi, Emanuel Aldea and Samir Bouaziz, Proceedings of the International Conference on Pattern Recognition (ICPR), 2016 final draft, bib
@inproceedings{nguyen16icpr,
 author    = {Dai Duong Nguyen and
               Abdelhafid Elouardi and
               Emanuel Aldea and
               Samir Bouaziz},
  title     = {{HOOFR:} An enhanced bio-inspired feature extractor},
  booktitle = {23rd International Conference on Pattern Recognition, {ICPR} 2016,
               Canc{\'{u}}n, Mexico, December 4-8, 2016},
  pages     = {2977--2982},
  year      = {2016}
}
SVM Classifier Fusion Using Belief Functions: Application to Hyperspectral Data Classification, Marie Lachaize, Sylvie Le Hégarat-Mascle, Emanuel Aldea, Aude Maitrot and Roger Reynaud, Proceedings of the 4th International Conference on Belief Functions (BELIEF), 2016 final draft, bib
@inproceedings{lachaize16belief,
 author    = {Marie Lachaize and
               Sylvie Le H{\'{e}}garat{-}Mascle and
               Emanuel Aldea and
               Aude Maitrot and
               Roger Reynaud},
  title     = {{SVM} Classifier Fusion Using Belief Functions: Application to Hyperspectral
               Data Classification},
  booktitle = {Belief Functions: Theory and Applications - 4th International Conference,
               {BELIEF} 2016, Prague, Czech Republic, September 21-23, 2016, Proceedings},
  pages     = {113--122},
  year      = {2016}
}
Robust crack detection for unmanned aerial vehicles inspection in an a-contrario decision framework, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Journal of Electronic Imaging, , 2015 final draft, bib
@article{aldea15jei,
author = {Aldea, Emanuel and Le H{\'e}garat-Mascle, Sylvie},
title = {Robust crack detection for unmanned aerial vehicles inspection in an a-contrario decision framework},
journal = {Journal of Electronic Imaging},
volume = {24},
number = {6},
pages = {061119},
year = {2015},
isbn = {1017-9909},
doi = {10.1117/1.JEI.24.6.061119},
URL = { http://dx.doi.org/10.1117/1.JEI.24.6.061119}
}
Robust crack detection strategies for aerial inspection, Emanuel Aldea and Sylvie Le Hégarat-Mascle, Proceedings of the Twelfth International Conference on Quality Control by Artificial Vision (QCAV), 2015 final draft, bib
@inproceedings{aldea15qcav,
author = {Aldea, Emanuel and Le H{\'e}garat-Mascle, Sylvie},
title = {Robust crack detection strategies for aerial inspection},
booktitle = {Proceedings of the Twelfth International Conference on Quality Control by Artificial Vision },
volume = {9534},
pages = {953413-953413-8},
year = {2015},
doi = {10.1117/12.2182920},
URL = { http://dx.doi.org/10.1117/12.2182920}
}
Spatio-Temporal Consistency for Head Detection in High-Density Scenes, Emanuel Aldea and Davide Marastoni and Khurom H. Kiyani, Proceedings of the ACCV Workshop on Human Identification for Surveillance (HIS), 2014 final draft, slides, bib
@inproceedings{aldea14accv1,
  author    = {Aldea, Emanuel  and Marastoni, Davide and Kiyani, Khurom H.},
  title     = {Spatio-Temporal Consistency for Head Detection in High-Density Scenes},
  booktitle = {Computer Vision - {ACCV} 2014 Workshops - Singapore, Singapore, November
               1-2, 2014, Revised Selected Papers, Part {III}},
  pages     = {665--679},
  year      = {2014}
}
Hybrid Focal Stereo Networks for Pattern Analysis in Homogeneous Scenes, Emanuel Aldea and Khurom H. Kiyani, Proceedings of the ACCV Workshop on Human Identification for Surveillance (HIS), 2014 final draft, slides, bib
@inproceedings{aldea14accv2,
author={Aldea, Emanuel and Kiyani, Khurom H.},
title     = {Hybrid Focal Stereo Networks for Pattern Analysis in Homogeneous Scenes},
booktitle = {Computer Vision - {ACCV} 2014 Workshops - Singapore, Singapore, November
               1-2, 2014, Revised Selected Papers, Part {III}},
pages     = {695--710},
year      = {2014}
}
SuperFAST: Model-based adaptive corner detection for scalable robotic vision, Gaspard Florentz and Emanuel Aldea, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014 final draft, bib
@inproceedings{florentz14iros, 
author={Florentz, Gaspard and Aldea, Emanuel}, 
booktitle={2014 IEEE/RSJ International Conference on Intelligent Robots and Systems}, 
title={Super{FAST}: Model-based adaptive corner detection for scalable robotic vision}, 
year={2014}, 
pages={1003-1010}, 
doi={10.1109/IROS.2014.6942681}, 
ISSN={2153-0858}, 
month={Sept}
}
SO(3)-invariant asymptotic observers for dense depth field estimation based on visual data and known camera motion, Nadège Zarrouati, Emanuel Aldea and Pierre Rouchon, Proceedings of the American Control Conference (ACC), 2012 final draft, bib
@inproceedings{zarrouati12acc,
author = {Zarrouati, Nad{\`e}ge and Aldea, Emanuel and Rouchon, Pierre}, 
title = {SO(3)-invariant asymptotic observers for dense depth field estimation based on visual data and known camera motion}, 
booktitle = {American Control Conference 2012}, 
address = {Montreal}, 
pages = {4116 - 4123}, 
year = {2012}
}
Robust depth regularization explicitly constrained by camera motion, Nadège Zarrouati, Emanuel Aldea and Pierre Rouchon, Proceedings of the International Conference on Pattern Recognition (ICPR), 2012 final draft, bib
@inproceedings{zarrouati12icpr,
author = {Zarrouati, Nad{\`e}ge and Aldea, Emanuel and Rouchon, Pierre}, 
title     = {Robust depth regularization explicitly constrained by camera motion},
booktitle = {Proceedings of the International Conference on Pattern Recognition (ICPR)},
year      = {2012},
pages     = {3606-3609}
}
Estimation dense de profondeur combinant approches variationnelles et observateurs asymptotiques, Nadège Zarrouati, Emanuel Aldea and Pierre Rouchon, Reconnaissance de Formes et Intelligence Artificielle (RFIA), 2012 final draft, bib
@inproceedings{zarrouati12rfia,
author = {Zarrouati, Nad{\`e}ge and Aldea, Emanuel and Rouchon, Pierre}, 
title     = {Estimation dense de profondeur combinant approches variationnelles et observateurs asymptotiques},
booktitle = {Actes de la conf{\'e}rence RFIA},
year      = {2012},
    pages = {978-2-9539515-2-3}
}
Object Tracking based on Particle Filtering with Multiple Appearance Models, Nicolas Widynski, Emanuel Aldea, Séverine Dubuisson and Isabelle Bloch, Proceedings of the International Conference on Computer Vision Theory and Applications (VISAPP), 2011 final draft, bib
@inproceedings{widynski11visapp,
  author = {Widynski, Nicolas and Aldea, Emanuel and Dubuisson, S{\'e}verine and Bloch, Isabelle},
  title = {Object Tracking based on Particle Filtering with Multiple Appearance Models},
  booktitle = {Proceedings of the  International Conference on Computer Vision Theory and Applications (VISAPP)},
  year = {2011},
  month = {March},
  pages = {604--609},
  address = {Algarve, Portugal}
}
Toward a Better Integration of Spatial Relations in Learning with Graphical Models, Emanuel Aldea and Isabelle Bloch, Advances in Knowledge Discovery and Management (AKDM), 2010 final draft, bib
@inproceedings{aldea10akdm,
  author = {Aldea, Emanuel and Bloch, Isabelle},
  title = {Toward a Better Integration of Spatial Relations in Learning with Graphical Models},
  bookTitle="Advances in Knowledge Discovery and Management",
year={2010},
publisher={Springer Berlin Heidelberg},
address={Berlin, Heidelberg},
pages={77--94},
isbn={978-3-642-00580-0},
doi={10.1007/978-3-642-00580-0_5},
url={http://dx.doi.org/10.1007/978-3-642-00580-0_5}
}
Vers une utilisation améliorée de relations spatiales pour l'apprentissage de données dans les modèles graphiques, Emanuel Aldea and Isabelle Bloch, Extraction et Gestion des Connaissances (EGC), 2009 final draft, bib
@inproceedings{aldea09egc,
  author = {Aldea, Emanuel and Bloch, Isabelle},
  title = {Vers une utilisation am{\'e}lior{\'e}e de relations spatiales pour l'apprentissage de donn{\'e}es dans les mod{\`e}les graphiques},
  booktitle = { Extraction et Gestion des Connaissances (EGC'2009)},
  year = {2009},
  pages = {271--282}
}
Kernel Fusion for Image Classification Using Fuzzy Structural Information, Emanuel Aldea, Geoffroy Fouquier, Jamal Atif and Isabelle Bloch, 3rd International Symposium on Visual Computing (ISVC), 2007 final draft, bib
@inproceedings{aldea07isvc,
  author = {Aldea, Emanuel and Fouquier, Geoffroy and Atif, Jamal and Bloch, Isabelle},
  title = {Kernel {F}usion for {I}mage {C}lassification {U}sing {F}uzzy {S}tructural {I}nformation},
  booktitle = {3rd International Symposium on Visual Computing ISVC07},
  address = {Lake Tahoe, USA},
  year = 2007,
  month = nov,
  volume = {LNCS 4842},
  pages = {307-317}
}
Classification d'images par fusion d'attributs flous de graphes, relations spatiales et noyaux marginalisés, Emanuel Aldea, Geoffroy Fouquier, Jamal Atif and Isabelle Bloch, Rencontres Francophones sur la Logique Floue et ses Applications (LFA), 2007 final draft, bib
@inproceedings{aldea07lfa,
  author = {Aldea, Emanuel and Fouquier, Geoffroy and Atif, Jamal and Bloch, Isabelle},
  title = {Classification d'images par fusion d'attributs flous de graphes, relations spatiales et noyaux marginalis{\'e}s},
  booktitle = {Rencontres Francophones sur la Logique Floue et ses Applications},
  year = 2007,
  pages = {25--32}
}
Image Classification using Marginalized Kernels for Graphs, Emanuel Aldea, Jamal Atif and Isabelle Bloch, 6th IAPR-TC15 Workshop on Graph-based Representations in Pattern Recognition (GBR), 2007 final draft, bib
@inproceedings{aldea07gbr,
  author = {Aldea, Emanuel and Atif, Jamal and Bloch, Isabelle},
  title = {Image {C}lassification using {M}arginalized {K}ernels for {G}raphs},
  booktitle = {6th IAPR-TC15 Workshop on Graph-based Representations in Pattern Recognition, GbR'07},
  address = {Alicante, Spain},
  year = 2007,
  month = jun,
  volume = {1},
  pages = {103--113}
}