Gabriel Dauphin,

Assistant Professor at L2TI,

at Institut Galilee at  Univerity Paris XIII

Phone : 33 61 49 40 28 29

Mail : gabriel.dauphin@univ-paris13.fr

 

 

Professional Experience

1998-2001 : PHD at ENST in Paris

2001-2002 : Teaching at University Paris XI

2002-         : Assistant Professor at L2TI

 

Teaching

Traitement d’image

    TP en Télécom 3 : IM\tp1_tin_L.pdf, IM\TP2_TIN_L.doc (IM\frequency.m, IM\huff2norm.m, IM\norm2huff.m), IM\TP3_TIN_L.doc (IM\daysailerGray.zip, IM\dailysailerRGB.zip)

    Cours en  Télécom 2 poly_tin.pdf (poly_tin_tab.pdf)

    TP en Télécom 2 :tp_tin_v2.pdf, (tp_tin_v2_tab.pdf, tp_tin_v2.zip)

   TD/TP en Master Images et Réseaux en M1 : IM\tp_tin_ir.pdf td_tin_m1ir_anc.pdf td_tin_m1ir.pdf (td_tin_m1ir_tab.pdf)

    Complément sur le codage Huffman : IM\TP_TIN_complement1.doc

Traitement du signal

   Cours TNS2010A.ppt, coursA.mp3, TNS2010B.ppt, coursB.mp3, TNS2010C.ppt, coursC.mp3, TNS2010D.ppt, coursD.mp3, TNS2010E.ppt, CoursE.mp3, TNS2010_1F.ppt, cours1F.mp3, TNS2010_2F.ppt, cours2F.mp3, TNS2010G.ppt, coursG.mp3, TNS2010H.ppt, coursH.mp3

   TD (td_tns.ps, poly.pdf, problemes_TNS.ps, , DevoirMaison.zip)

   QCM Un programme réalisant un QCM est aussi disponible en téléchargement (QUIZ_APP.zip, QUIZ_APP.exe) [Aymen Graba pour la conception et l’implémentation], ou sur Web (QUIZ_WEB.html) [Kheloudja Fettouche, Ameni Letaief, Yasmine Ntsiantou Owaye, Adama Sissokho] et une nouvelle version qcm\index.php[Harouna Ba, Sarah Essqali]

  TP simulation du signal 2sin(2*pi*f0*t+pi/2) échantillonné à fe en temps réel lent

(complement_tns_tp1.pdf, complement_tns_tp1.zip)

Programmation en Matlab

   TP (exercice_enonce.pdf examen_MatLab_I.pdf, examen_MatLab_Ja.pdf, examen_Matlab_Jb.pdf),

   Cours (d'après un document de C. Kulcsar) polyMatlabCpl.pdf (polyMatlabCplTablette.pdf)

Programmation en C

  Cours te2_td_HarC.pdf (te2_td_HarC_tab.pdf)

 

Pour lire les fichiers écrits en postscript et avec l’extension .ps : télécharger Ghostscript puis GSview sur http://pages.cs.wisc.edu/~ghost/.

L'extension tab signifie que le fichier peut être a priori être lu sur une grande tablette

 

 

First Research Domain

In most of the existing image quality assessment approaches, the main objective is to develop measures that are consistent with the subjective evaluation. Therefore, the performance of a given image quality metric is evaluated against the MOS determined from a series of subjective tests performed on a database. A plethora of image quality metrics has been developed. However, a few studies have been reported on the analysis and comparison of these metrics. The research aims at finding invariance laws with which one may test image quality metrics. The oral presentation [1] shows a new viewpoint. Paper [2] focuses on one specific property, namely the sensitivity to background luminance. It shows on synthetic images that Weber’s law is not consistent with human visual system (HVS), a photometric invariance principle is proposed and an algorithm is derived to test whether a given quality metric complies with Weber’s law or with that specific property of HVS (qualityMetric2Alpha.m, psnr.m, ssim.m, ssim_index.m,  vdp.m). To what extent Weber’s law is consistent with HVS will be further investigated.

 

Second Research Domain

Context:  object recognition or gate recognition can be interpreted as a classification of images or of video (i.e. images or videos containing the object or the event as opposed to those not containing it). Our contribution is to take into account the specificity of the application and of the resolution of the camera.

Application in Focus : improve home medical care by setting cameras. Real-time processing of videos could enable to detect faintness or repeated medication intakes and to trigger a voice message or a call to a medical advisor.

L’installation d’une caméra à domicile pourrait redonner une vie autonome dans leur domicile à des personnes âgées qui normalement auraient été contraintes à un placement en institution spécialisée. L’analyse en temps réel des séquences vidéos pourraient en effet permettre de détecter l’apparition d’un malaise ou de compter le nombre de prises de médicaments afin d’initier un dialogue avec le patient ou de déclencher l’alerte.

Tracking Example : The following video shows an example of drinking glass tracking: Intaking Medication.avi

Contribution : [3] shows a contour estimation algorithm specific to low-resolution camera. [4] shows how location and movement of interest points are in themselves informative.

 

Third Research Domain

Context: With the recent advances in stereoscopic display technologies, there could be a growing demand for designing efficient stereo image compression techniques. We are trying to improve the coding performance by modifying the disparity estimation process (disparity is this varying shift that object have on the left view as compared to that in the right views, it is this shift that yields a perception of depth).

Contribution: [5] builds a dictionary that links together disparities yielding similar reconstructions.

 

List of  Publications

      A. Kadaikar, G. Dauphin, A. Mokraoui, Sequential Block-Based Disparity Map Estimation Algorithm for Stereoscopic Image Coding. Elsevier Signal Processing : Image Communication, September 2015, DOI 10.1016/j.image.2015.09.007.

[5]  G. Dauphin, M. Kaaniche, A. Mokraoui, Block dependent dictionary based disparity compensation for stereo image coding, IEEE International Conference on Image Processing, ICIP 2015, 27-30 September Québec City Canada.

  Dauphin15_paper.pdf Dauphin15_poster.pdf Dauphin15_lecture.wma

      A. Kadaikar, G. Dauphin, A. Mokraoui, Estimation de cartes de disparité basée sur l’algorithme revisité d’appariement de blocs pour la compression d’images stéréoscopiques, In Proceding of the GRETSI Conferences, 8-11 septembre 2015, Lyon, France.

      A. Kadaikar, G. Dauphin, A. Mokraoui, Méthode d'optimisation pour l'appariement de pixels d'images stéréoscopiques basée sur une métrique conjointe entropie-distorsion, Compression et Représentation des Signaux Audiovisuels, CORESA, 26-27 Novembre 2014, Reims.

      A. Kadaikar, A. Mokraoui , G. Dauphin, Entropy-constrained Dense Disparity Map Estimation Algorithm for Stereoscopic Images, European Signal Processing Conference, EUSIPCO, 1-5 September 2014, Lisbon Portugal.

[4] G. Dauphin, Tracking Based Solely on Location of Interest Points. In Proceeding of 4th European Workshop on Visual Information Processing. 6 pages, Paris, France, June 2013.

      Dauphin13_paper.pdf, Dauphin13_poster.pdf, Dauphin13_demo.m

[3] G. Dauphin and S. Khanfir, Background Suppression with Low-Resolution Camera in the Context of Medication Intake Monitoring. In Proceeding of 3rd European Workshop on Visual Information Processing. 6 pages, Paris, France, July 2011. EUVIP'11.

      Dauphin11_paper.pdf, Dauphin11_poster.pdf

[2] G. Dauphin and P. Viaris de Lesegno, Assessment of Quality Metrics with Reference Based on Uniform Colour Spaces. In Proceeding of 2nd European Workshop on Visual Information Processing. 6 pages, Paris, France, July 2010. EUVIP'10.

      Dauphin10_paper.pdf Dauphin10_lecture.mp3

[1]  G. Dauphin, A. Beghdadi, A. Bouzerdoum, HVS-Inspired Image Processing: An Overview, In Proceeding of the 1st International Workshop on Visual Signal Processing and its Applications, Université Paris 13, 12-13 November 2008. VISPA'08.

      Dauphin08_slides.pdf

      A. Beghdadi, G. Dauphin, A. Bouzerdoum, Image Analysis Using Local Band Directional Contrast. In Proceeding of the International Symposium on Intelligent Multimedia, Video and Speech Processing, Hong Kong, October 20-22, 2004, invited. ISIMP'04.

      G. Dauphin, A. Beghdadi and P. Viaris de Lesegno, A Local Directional Bandlimited Contrast, In Proceeding of International Symposium on Signal Processing and its Applications, 4 pages, Paris, France, 2003, ISSPA'03.

      P. Bonnin, O. Stasse, V. Hugel, P. Blazevic, G. Dauphin, Towards a Method to Compare and to Evaluate Fast Pixel Gathering Mechanisms For Real Time Robotic Vision Systems. In Proceeding of International Symposium on Signal Processing and its Applications, 4 pages, Paris, France, 2003, ISSPA'03.

      G. Dauphin and D. Matignon. Positivity and Dissipativity of Oscillating Diffusive Filters, Application to the Stability of Coupled Systems. In Proceeding of Mathematical Theory of Networks and Systems symposium, 10 pages, Notre Dame, Indiana, USA, 2002. MTNS'02.

      G. Dauphin and D. Matignon. Application of Diffusive Representations to Discretized Fractional Systems: Stability, Positivity and Simulations Issues. In Proceeding of European Control Conference, pages 1490-1495, Porto, Portugal, September 2001. ECC'01.

      G. Dauphin, D. Heleschewitz and D. Matignon. Extended Diffusive Representations and Application to Non-Standard Oscillators. In Proceeding of Mathematical Theory of Networks and Systems symposium, 10 pages, Perpignan, France, June 2000. MTNS'00.

      G. Dauphin and D. Matignon. Premiers résultats sur les représentations diffusives à temps discret et application aux filtres ARFIMA. In Proceeding of Journées Doctorales d'Automatique, pages 45-48, Nancy, France, septembre 1999. GDR AUTOMATIQUE.

      G. Dauphin and D. Matignon. Stabilité interne d'un système couplé à temps discret, formé d'un oscillateur et d'un filtre diffusif dissipatif. In Journéées Doctorales d'Automatique, pages 189-194, Toulouse, France, septembre 2001. GDR AUTOMATIQUE.