Scientific Staff Member (BSIK/BRICKS scholar);
Centrum voor Wiskunde en Informatica ( CWI ) 
Group: Semantics from Sensors (formerly PNA 4 - Signals and Images)
Science Park 123
1098 XG Amsterdam, The Netherlands
email: a.a.salah_at_cwi.nl

Upcoming Events

• Visit to INRIA 28 April, Bordeaux.
• Int. Conference on Computational Sustainability, 8-11 June, Ithaca.
• European Conference on Computing and Philosophy, 2-4 July, Barcelona.
• Annual Meeting of the Cognitive Science Society, 29 July - 1 August, Amsterdam.

Education

• 2007 PhD Boğazici University Computer Engineering Department
• 2000 MS Boğazici University Computer Engineering Department
• 1998 BS Boğazici University Computer Engineering Department
• 1994 İstanbul Özel Alman Lisesi/Deutsche Schule Istanbul

Scientific Interests

• Pattern recognition and machine learning
• Biometrics - 3D face recognition, fusion of multiple biometrics
• Natural and artificial vision, selective attention
• Ambient intelligence, sensor networks
• Unsupervised learning and feature extraction
• Cognitive science and philosophy of mind
• Semiotics and biosemiotics

Journal Papers

• Poh, N., T. Bourlai, J. Kittler, L. Allano, F. Alonso, O. Ambekar, J. Baker, B. Dorizzi, O. Fatukasi, J. Fierrez, H. Ganster, J.-O. Garcia, D. Maurer, A.A. Salah, T. Scheidat, C. Vielhauer, "Benchmarking Quality-Dependent and Cost-sensitive Multimodal Biometric Fusion Algorithms," accepted for publication, IEEE Trans. Information Forensics and Security.
• Salah, A.A., R. Morros, J. Luque, C. Segura, J. Hernando, O. Ambekar, B. Schouten, E.J. Pauwels, "Multimodal identification and localization of users in a smart environment," Journal on Multimodal User Interfaces, available online, 2008.
• Akdağ Salah, A.A., A.A. Salah, "Technoscience Art: A Bridge between Neuroesthetics and Art History?," Review of General Psychology, vol 12(2), pp.147-158, 2008.
• Salah, A.A., N. Alyüz, L. Akarun, "Registration of 3D Face Scans with Average Face Models", Journal of Electronic Imaging, January-March 2008.
• Salah, A.A., H. Çınar, L. Akarun, B. Sankur, "Robust Facial Landmarking for Registration", Annals of Telecommunications, vol.62, no.1-2, pp.1608-1633, 2007.
• Salah, A.A., E. Alpaydın, L. Akarun, "A Selective Attention Based Method for Visual Pattern Recognition with Application to Handwritten Digit Recognition and Face Recognition," IEEE Trans. Pattern Analysis and Machine Intelligence, Vol.24, No.3, pp. 420-425, March 2002.
  (Fig 1 erroneous in pdf; get the correct Fig here)

Book Chapters

• Salah, A.A., "Machine Learning for Biometrics," E. Soria, J.D. Martín, R. Magdalena, M.Martínez, A.J. Serrano (eds.), Handbook of Research on Machine Learning Applications, in press.
• Gökberk, B., A.A. Salah, N. Alyüz, L. Akarun, "3D Face Recognition: Technology and Applications," M. Tistarelli, S.Z. Li, R. Chellappa (eds.) Biometrics for Surveillance and Security, Springer-Verlag, London, in press.
• Salah, A.A., "Perceptual Fusion in Humans and Machines" in O. Tanrıdağ (ed.), "Cognitive Neuroscience at Marmaris - An Interdisciplinary Book on Selected Themes From The Previous Meetings 2005-2007," Pharmapublication Planning, Istanbul, 2008.
• Gökberk, B., A.A. Salah, L. Akarun, R. Etheve, D. Riccio, J.-L. Dugelay, "3D Face Recognition" in D. Petrovska-Delacrétaz, G. Chollet, B. Dorizzi (eds.), "Guide to Biometric Reference Systems and Performance Evaluation," Springer-Verlag, London, 2008.
• Tistarelli, M., M. Bicego, J.L. Alba-Castro, D. Gonzàlez-Jimènez, A. Mellakh, A.A. Salah, D. Petrovska-Delacrètaz, B. Dorizzi, "2D Face Recognition" in D. Petrovska-Delacrétaz, G. Chollet, B. Dorizzi (eds.), "Guide to Biometric Reference Systems and Performance Evaluation," Springer-Verlag, London, 2008.

International Conference and Workshop Papers

• Salah, A.A., B. Schouten, "Semiosis and the relevance of context for the AmI environment", accepted for 9th European Conference on Computing and Philosophy, Barcelona, 2009.
• Yücel, Z., A.A. Salah, "Head pose and neural network based gaze direction estimation for joint attention modeling in embodied agents," accepted for 31st Annual Conference of Cognitive Science Society, Amsterdam, 2009.
• Poh, N., C.H. Chan, J. Kittler, S. Marcel, C. Mc Cool, E. Argones Rua, J.L. Alba Castro, M. Villegas, R. Paredes, V. Struc, N. Pavesic, A.A. Salah, H. Fang, N. Costen, "Face Video Competition," accepted for The 3rd IAPR/IEEE Int. Conf. on Biometrics (ICB'09), Alghero, 2009.
• Dibeklioğlu, H., A.A. Salah, L. Akarun, "3D Facial Landmarking under Expression, Pose, and Occlusion Variations," IEEE Second International Conference on Biometrics: Theory, Applications and Systems (BTAS'08), Washington, Sept. 29-Oct.1, 2008.
• Schouten, B., A.A. Salah, "Empowering the End-user in Biometrics," 10th IEEE Int. Conf. on Control, Automation, Robotics and Vision (ICARCV'08), Hanoi, 17-20 December, 2008.
• Alyüz, N., B. Gökberk, H. Dibeklioğlu, A. Savran, A.A. Salah, L. Akarun, B. Sankur, "3D Face Recognition Benchmarks on the Bosphorus Database with Focus on Facial Expressions" in Proc. First European Workshop on Biometrics and Identity Management (BIOID'08), Roskilde, 7-9 May, 2008.
• Tavenard, R., A.A. Salah, E.J. Pauwels, "Searching for Temporal Patterns in AmI Sensor Data," in M. Muhlhauser, A. Ferscha, E. Aitenbichler (eds.), "Constructing Ambient Intelligence: AmI-07 Workshops Proceedings", LNCS, Darmstadt, Germany, 7-10 November, 2007.
• Pauwels, E., A.A. Salah, R. Tavenard, "Sensor Networks for Ambient Intelligence," Workshop on Multimedia Signal Processing (MMSP), Crete, October 2007.
• Bicego, M., A.A. Salah, E. Grosso, M. Tistarelli, L. Akarun,"Generalization in Holistic versus Analytic Processing of Faces," 14th Int. Conf. on Image Analysis and Processing, Modena, Italy, 10-14 September, 2007.
• Morros, R., A.A. Salah, B. Schouten, C.S. Perales, J.L. Serrano, O. Ambekar, C. Kayalar, C. Keskin, L. Akarun, "Event recognition for meaningful human-computer interaction in a smart environment," Proc. 3rd eNTERFACE Workshop, pp. 71-86, Istanbul, August 2007.
• Salah, A.A., N. Alyüz, L. Akarun, "Alternative face models for 3D face registration," in SPIE Conf. on Electronic Imaging, Vision Geometry, San Jose, January 2007.
• Salah, A.A., M. Bicego, L. Akarun, E. Grosso, M. Tistarelli, "Hidden Markov Model-based face recognition using selective attention," in SPIE Conf. on Human Vision and Electronic Imaging, San Jose, January 2007.
• Salah, A.A., L. Akarun, "3D Facial Feature Localization for Registration," International Workshop on Multimedia Content Representation, Classification and Security (MRCS’06), September 2006.
• Salah, A.A., B. Gökberk, L. Akarun, J.R. Tena, M. Hamouz, J. Kittler, M. Bicego, E. Grosso, M. Tistarelli, "3D face verification using shape and texture/3D face registration and landmark localization," Biosecure Workshop, Vigo, September 2006.
• Salah, A.A., L. Akarun, "Factor Analysis Based Robust 3D Facial Landmarking", 3. Summer School for Advanced Studies on Biometrics for Secure Authentication: New Sensors, Evaluation and Database Collection, Alghero, June 2006.
• Çınar Akakın, H., A.A. Salah, L. Akarun, B. Sankur, "2D/3D facial feature extraction", in Proc. SPIE Conf. on Electronic Imaging, San Jose, 2006.
• Akarun, L., B. Gökberk, A.A. Salah, "3D Face Recognition for Biometric Applications", in Proc. European Signal Processing Conference, Antalya, 2005.
• Gökberk, B., A.A. Salah, L. Akarun, "Rank-based Decision Fusion for 3D Shape-based Face Recognition," in Takeo Kanade, Anil Jain, Nalini K. Ratha (eds.) Lecture Notes in Computer Science, Volume 3546/2005 Audio- and Video-Based Biometric Person Authentication: 5th International Conference, AVBPA 2005.
• Salah, A.A., E. Alpaydın, "Incremental Mixtures of Factor Analyzers," in Proc. International Conference on Pattern Recognition (ICPR04), vol.1, pp.276-279, Cambridge, 2004.
• Salah, A.A., S. Redman, G. Kovacs, "Clustering under Multiobjective Optimization Constraints with Genetic Algorithms" in Proc. of the Santa Fe Complex Systems Summer School, Santa Fe, NM: SFI, 2002.
• Labiouse, C., A.A. Salah, I. Starikova, "The Impact of Connectivity on the Memory Capacity and the Retrieval Dynamics of Hopfield-type Networks" in Proc. of the Santa Fe Complex Systems Summer School, Santa Fe, NM: SFI, pp.77-84, 2002.
• Banerjee, N., Z. Dezso, V. Kolyadin, A. Marhan, A.A. Salah, "Lunch Network: A Glimpse of Social Dynamics of SFI Students" in Proc. of the Santa Fe Complex Systems Summer School, Santa Fe, NM: SFI, 2002.
• Salah, A.A., E. Alpaydın, L. Akarun, "A Selective Attention Based Method for Visual Pattern Recognition," in Moore, J.D., and Stenning, K. (eds.) Proc. 23rd Annual Conference of Cognitive Science Society, pp. 881-886, Edinburgh, August 2001.
• Salah, A.A., E. Alpaydın, L. Akarun, "A Selective Attention Based Optical Recognition of Handwritten Digits," Proceedings of the First IEEE Balkan Conference on Signal Processing, Communications, Circuits and Systems, 2000.

Local Publications

• Dibeklioğlu, H., A.A. Salah, L. Akarun, "Expression, Pose and Occlusion Resistant 3D Facial Landmarking," IEEE SİU 2009, Antalya, 2009. (Recipient of Alper Atalay 2nd best student paper award)
• Salah, A.A., "Face Authentication for an Activity-Related Biometrics System," IEEE SİU 2009 Antalya, 2009.
• Pauwels, E., A.A. Salah, P. de Zeeuw, Tackling the Semantic Gap in Sensor Networks, "ERCIM News, Number 76, January 2009.
• Salah, A.A., "A Needle in the Brain," ERCIM News, Number 73, April 2008.
• Salah, A.A., E.J. Pauwels, "Prediction and Voronoi graph construction with a dense network of simple sensors", IEEE SİU 2008, 20-22 April 2008, Didim, Turkey.
• Salah, A.A., A.P. Saygın, "Retinotopy and Selective Visual Attention in Humans and Computers", IEEE SİU 2008, 20-22 April 2008, Didim, Turkey.
• Salah, A.A., "Algısal Tümleştirme", Dördüncü Bilgi İşleyen Makina Olarak Beyin Sempozyumu, 4-5 Nisan 2008, Albert Long Hall, Boğaziçi Üniversitesi.
• Salah, A.A., "Perceptual Information Fusion in Humans and Machines," Cognitive Neuroscience Forum, vol.2, 2007.
• Alyüz, N., Salah, A.A., L. Akarun, "Alternative average face models for 3D face registration," IEEE SİU 2007, 11-13 June, Eskişehir, Turkey.
• Salah, A.A., "İnsan ve Bilgisayarda Yüz Tanıma," Uluslararası Kognitif Nörobilim Sempozyumu, Marmaris, Turkey, Mayıs 2006.
• Salah, A.A., L. Akarun, "Gabor Factor Analysis for 2D+3D Facial Landmark Localization," IEEE SİU 2006, 17-19 Nisan 2006, Antalya, Turkey.
• Salah, A.A., B. Gökberk, L. Akarun, "3D Face Recognition", Proceedings of the Fifth GAP Engineering Congress, 26-28 April 2006, Şanlıurfa, Turkey.
• Salah, A.A., "İnsan ve Bilgisayarda Yüz Tanıma", Üçüncü Bilgi İşleyen Makina Olarak Beyin Sempozyumu, 24 Aralık 2005, Albert Long Hall, Boğaziçi Üniversitesi.
• Gökberk, B., A.A. Salah, L. Akarun, "Üç Boyutlu Şekil Bilgisiyle Yüz Tanımada Sıralamaya Dayalı Karar Birleştirimi", IEEE SİU 2005.
• Salah, A.A., E. Alpaydın, "Incremental Mixtures of Factor Analyzers," Technical Report, FBE/CMPE-01/2004-7, Department of Computer Engineering, Boğaziçi University, 2004. (This is the longer version of the ICPR'04 paper).
• Salah, A.A., E. Alpaydın, L. Akarun, "Selective Attention Based Pattern Recognition," Technical Report, FBE/CMPE-03/2001-12, Department of Computer Enginnering, Boğaziçi University, 2001.
• Salah, A.A., A Selective Attention Based Method for Optical Handwritten Digit Recognition (M.S. Thesis - 2000)
• Salah, A.A., Collective Intelligence, Panel Discussion at the 7th International Istanbul Biennial, September 23, 2001 with Peter Gardenfors, Kazuhiko Hachiya, and Rafael Lozano-Hemmer. (download talk in DOC format/Teknotürk'te çıkan yazı için tıklayın: İstanbul Bienali'nde Bir Panel: Kolektif Zeka)
• Akın, L., A.A. Salah, Boğaziçi Üniversitesi'ndeki Robot Araştırmaları - Geleceğe Giden Yol, Teknotürk.
• Salah, A.A., E. Alpaydın, L. Akarun, "A Selective Attention Based Recognition Framework with Application to Handwritten Numerals," Proceedings of the Ninth Turkish Symposium on Artificial Intelligence and Neural Networks (TAINN), 2000.
• Salah, A.A., "An Evolutionary Algorithm Tool in Smalltalk" (B.S. Thesis - 1998)

PhD Dissertation: Biologically Motivated 3D Face Recognition

Face recognition has been an active area of study for both computer vision and image processing communities, not only for biometrics but also for human-computer interaction applications. The purpose of the present work is to evaluate the existing 3D face recognition techniques and seek biologically motivated methods to improve them. We especially look at findings in psychophysics and cognitive science for insights. We propose a biologically motivated computational model, and focus on the earlier stages of the model, whose performance is critical for the later stages. Our emphasis is on automatic localization of facial features. We first propose a strong unsupervised learning algorithm for flexible and automatic training of Gaussian mixture models and use it in a novel feature-based algorithm for facial fiducial point localization. We also propose a novel structural correction algorithm to evaluate the quality of landmarking and to localize fiducial points under adverse conditions. We test the effects of automatic landmarking under rigid and non-rigid registration methods. For the rigid registration approach, we implement the iterative closest point method (ICP). The most important drawback of ICP is the computational cost of registering a test scan to each scan in the gallery. By using an average face model in rigid registration, we show that the computation bottleneck can be eliminated. Following psychophysical arguments on the "other race effect", we reason that organizing faces into different gender and morphological groups will help us in designing more discriminative classifiers. We test this claim by employing different average face models for dense registration. We propose a shape-based clustering approach that assigns faces into groups with nondescript gender and race. Finally, we propose a regular re-sampling step that increases the speed and the accuracy significantly. These components make up a full 3D face recognition system.

This work was the recipient of the European Biometrics Forum (EBF) 2006 Biometrics Research Award (More on that on EBF portal and Biometrics Technology Today). It was supported by the following projects:

• 3D Object Recognition, The Scientific and Technological Research Council of Turkey (TUBITAK), Project Code: 104E080
• Perceptual Human Computer Interaction, T.R. Prime Ministry State Planning Organization (DPT), Project Code: 03K120250
• BIOSECURE FP6 network of excellence project on multimodal biometrics.

Master Thesis - Selective Attention Based Pattern Recognition

The human visual mechanism, which develops in parallel with the brain, is physically very complex and functions in a very efficient manner. Human vision is mainly a serial process, in that we do not absorb the whole information content of the visual field at once, but selectively attend to locations that have low probability, and thus contain more information. Our aim is to model this process in a biologically plausible framework to achieve some of the desired properties of natural visual systems.

Code

• The IMoFA algorithm (incremental mixtures of factor analyzers) in MATLAB (download). This algorithm performs automatic clustering and model selection in an unsupervised manner. The model is basically a Gaussian mixture, but from a complexity point of view, the output will be between a diagonal-covariance GMM and a full-covariance GMM, thanks to the latent variable assumptions. The demonstration code models the Iris data with IMoFA, and shows how classification can be effected with the trained models.
• The GOLLUM algorithm (Gaussian outlier localization with likelihood margins) in MATLAB (download). GOLLUM is a morphological algorithm that learns landmark distributions for the purpose of detecting and correcting a given landmark constellation, irrespective of the underlying feature distribution. It can be used for detecting landmark localization errors automatically, providing expected locations for landmarks, and for correcting a given landmark location based on other landmarks. The provided code uses the Levenberg-Marquard algorithm from the Optimization Toolbox of MATLAB, but this step can be replaced with a custom algorithm. The demonstration code shows how to operate GOLLUM to correct randomly corrupted facial landmark data.
• A Genetic Algorithm Package for Dolphin Smalltalk (download)
• Latrunculli Game in SWI-Prolog (contains a stand-alone genetic algorithm module)

Summer and Spring Schools, Workshops

• The eNTERFACE Workshop, LIMSI, Orsay, 4-29 August, 2008.
• Study Group Mathematics with Industry 2008, Twente, 28 January - 1 February, 2008.
• The eNTERFACE Workshop, Boğaziçi University, Istanbul, July 16-August 10, 2007.
• Biosecure Workshop in Vigo, Spain, 13-15 September, 2006.
• Biologically Inspired Information Fusion Workshop, Surrey, 22-23 August, 2006.
• Geometric Morphometrics Workshop, Ankara, June 12-16, 2006.
• Summer School for Advanced Studies on Biometrics for Secure Authentication: New Sensors, Evaluation and Database Collection, Alghero, June 5-9, 2006.
• Santa Fe Institute Complex Systems Summer School, Budapest, July 8 to August 2, 2002.
• 8th International Summer School on Cognitive Science at New Bulgarian University, Sofia, July 9-30, 2001.
• BISC European School on Intelligent Data Analysis, Palermo, March 26-30, 2001.
• 7th International Summer School on Cognitive Science at New Bulgarian University, Sofia, July 10-30, 2000.

Invitation to Cognitive Science

Would you be interested in a Cognitive Science program in Boğaziçi University? Visit the M.A. Program in Cognitive Science webpage.

Personal Information

• Viktor Sina is on Facebook, Picasa and Vimeo
• Resume (March 2008)
• BiBTeX file for publications
• Ex-Member of the Amphion Choir
• Ex-Member of the Perceptual Intelligence (PI) Lab
• Ex-Editor, Teknotürk
• Ex-Member of the Chaos & Complexity Group at Boğaziçi University
• Ex-Member of the Istanbul European Choir

Last Updated: April 2009