s 

Dimitris Samaras Assistant Professor, Director Image Analysis Lab, Computer Science Department 2429 Computer Science Stony Brook University Stony Brook, NY 11794-4400 (631) 632-8464 samaras AT cs DOT sunysb DOT edu

RESEARCH CURRICULUM VITAE TEACHING PUBLICATIONS RESEARCH GROUP

Education

· Ph.D. in Computer Science, January 2001.University of Pennsylvania,  (My old webpage)

· M.S in Computer Science, June 1994. Northeastern University.

· Diploma in Computer Engineering and Informatics, June 1992. University of Patras, Greece.

Teaching

· Available Projects for  CSE523/524 UPDATED Jan 2005

· Ongoing:   CSE656 Seminar in Computer Vision (graduate)

· Fall 2005: CSE 378/525  Introduction to Robotics   (undergraduate/graduate)

· Spring 2002-5: CSE527 Introduction to Computer Vision (graduate)

· Fall 2004:  CSE 601/ESE559  (with G Gindi and J Liang) Advanced Image Processing  (graduate)

· Fall 2004:        CSE 592  Introduction to Robotics   (graduate)

· Fall 2004:        CSE 681  (with T Pavlidis) Topics in Computer Vision   (graduate)

· Fall 2003:        CSE 615  Advanced Image Analysis  (graduate)

· Fall 2001-3:     CSE 390  Introduction to Visual Computing  (undergraduate)

· Spring 2001:   CSE527/627 Introduction to Computer Vision (graduate)

· Fall 2000:        ESE 358/CSE 327  Computer Vision  (undergraduate)

Research Interests

Computer vision, computer graphics, machine learning, medical imaging, animation and simulation, image based rendering, physics-based modeling.

My research up to now has focused on explaining visual data for Computer Vision, Computer Graphics and Medical Image Analysis, through the appropriate physical and statistical models. A central interest is in modeling the interaction of 3D shape and illumination, (a major source of variability in images) for applications such as shape and motion estimation, object recognition and augmented reality. Further interest in 3D shape deformation is fueled by the availability of image and range data that allows statistical modeling of non-rigid motion. The construction of such statistical models leads to the problem of accurate matching of 3D data. A natural application area for all the above questions is the field of human modeling and especially faces. We have been focusing on topics such as facial appearance under variable illumination and facial expression modeling for biometrics and human computer interaction. My general interest in human modeling has led to exciting collaborations with psychologists who collect visual data about human behavior through multiple modalities such as eye-trackers and fMRI bran imaging. Thus based on recent neuropsychological findings we are exploring the application of machine learning techniques to the analysis of brain images. Current sources of funding include NSF, NIH (NIDA), DoE and DoJ

Selected Projects

Facial Modeling & Analysis · Expression Modeling · Face Recognition

Machine Learning for the Brain

Illumination Modeling for Computer Vision & Computer Graphics

Shape Modeling & Analysis

Publications

2006

· Face Recognition from A Single Training Image under Arbitrary Unknown Lighting using Spherical Harmonics,
Lei Zhang, Dimitris Samaras, In IEEE Trans on Pattern Analysis and Machine Intelligence (to appear)

· Integration of Deformable Contours and a Multiple Hypotheses isher Color Model for Robust Tracking in Varying Illuminant Environments,
Francesc Moreno-Noguer, Alberto Sanfeliu, Dimitris Samaras In Image and Vision Computing (to appear)

2005

· A Computational Model of Eye Movements during Object Class Detection,
Wei Zhang, Dimitris Samaras, Hyejin Yang, Greg Zelinsky. In NIPS, 2005 (to appear).

· The Role of Top-down and Bottom-up Processes in Guiding Eye Movements during Visual Search,
Greg Zelinsky, Wei Zhang, Bing Yu, Xin Chen, Dimitris Samaras, In NIPS, 2005 (to appear).

· Modeling Neuronal Interactivity using Dynamic Bayesian Networks,
Lei Zhang, Dimitris Samaras, Nelly Alia-Klein, Nora Volkow, Rita Goldstein, In NIPS 2005 (to appear).

· Exploiting Temporal Information in Functional Magnetic Resonance Imaging Brain Data,
Lei Zhang, Dimitris Samaras, Nelly Klein, Dardo Tomasi, Lisa Cottone, Andreana Leskovjan, Nora Volkow, Rita Goldstein, In Proc MICCAI 2005, pp. 679-687.

· High Resolution Tracking of Non-Rigid 3D Motion of Densely Sampled Data Using Harmonic Maps,
Yang Wang, Mohit Gupta, Song Zhang, Sen Wang, David Gu, Dimitris Samaras, Peisen Huang, In Proc. of ICCV 2005, pp. 388-395.

· Integration of Conditionally Dependent Object Features for Robust Figure/Background Segmentation,
Francesc Moreno-Noguer, Alberto Sanfeliu, Dimitris Samaras. In Proc. of ICCV 2005 pp. 1713-1720.

· Face Reconstruction across Different Poses and Arbitrary Illumination Conditions,
Sen Wang, Lei Zhang, Dimitris Samaras, In Proc AVBPA 2005, pp. 91-101.

· Object class recognition using multiple layer boosting with multiple features.
Wei Zhang, Bing Yu, Greg Zelinsky, and Dimitris Samaras. In IEEE Proc. of CVPR 2005, pp II:323-330. Background set (Download)

· Face Synthesis and Recognition under Arbitrary Unknown Lighting using a Spherical Harmonic Basis Morphable Model,
Lei Zhang, Sen Wang, Dimitris Samaras In IEEE Proc. of CVPR 2005 pp. II:209-216

· Machine Learning for Clinical Diagnosis from Functional Magnetic Resonance Imaging,
Lei Zhang, Dimitris Samaras, Dardo Tomasi, Nora Volkow Rita Goldstein, In IEEE Proc. of CVPR 2005 pp. I:1211 – 1217.

· Image-Driven Re-targeting and Relighting of Facial Expressions,
Lei Zhang, Yang Wang, Sen Wang, Dimitris Samaras, In Proc CGI 2005, pp. 11-18.

2004

· High Resolution Acquisition, Learning and Transfer of Dynamic 3-D Facial Expressions
Yang Wang, Xiaolei Huang, Chan-Su Lee, Song Zhang, Zhiguo Li, Dimitris Samaras, Dimitris Metaxas, Ahmed Elgammal, Peisen Huang In Computer Graphics Forum (EuroGraphics 2004) pp. III: 677-686.

· Fusion of a Multiple Hypotheses Color Model and Deformable Contours for Figure Ground Segmentation in Dynamic Environments
Francesc Moreno-Noguer, Alberto Sanfeliu, Dimitris Samaras In Proc. 3rd IEEE Workshop on Articulated and Nonrigid Motion, ANM’04, (in conjunction with CVPR’04).

· A Hierarchical Framework for High Resolution Facial Expression Tracking
Xiaolei Huang, Song Zhang, Yang Wang, Dimitris Metaxas, Dimitris Samaras In Proc. 3rd IEEE Workshop on Articulated and Nonrigid Motion, ANM’04, (in conjunction with CVPR’04).

· Pose Invariant Face Recognition under Arbitrary Unknown Lighting using Spherical Harmonics,
Lei Zhang, Dimitris Samaras. In Proc Biometric Authentication Workshop 2004, (in conjunction with ECCV2004)

· Shape Reconstruction from 3D and 2D Data Using PDE-Based Deformable Surfaces,
Ye Duan, Liu Yang, Hong Qin, Dimitris Samaras In Proc ECCV 2004, pp III:238-251

2003

· Multiple Directional Illuminant Estimation from a Single Image.
Yang Wang, Dimitris Samaras In Proc. IEEE Workshop on Color and Photometric Methods in Computer Vision (in conjunction with ICCV 2003).

· Estimation of Multiple Directional Light Sources for Synthesis of Augmented Reality Images.
Yang Wang, Dimitris Samaras In Graphical Models, July 2003, pp. 185-205. (Special Issue on Pacific Graphics 2002).

· Face Recognition Under Variable Lighting using Harmonic Image Exemplars.
Lei Zhang, Dimitris Samaras. In Proc. CVPR 2003 pp I: 19-25.

· Using Multiple Cues for Hand Tracking and Model Refinement
Shan Lu, Dimitris Metaxas, Dimitris Samaras, John Oliensis. In Proc. CVPR 2003 pp. II:443-450.

· Illumination Constraints in Deformable Models for Shape and Light Direction Estimation.
Dimitrios Samaras, Dimitris Metaxas.  In IEEE Trans. PAMI Feb. 2003, pp.  247-264

2002

· Model-based Integration of Visual Cues for Hand Tracking.
Shan Lu, Gang Huang , Dimitris Samaras, Dimitris Metaxas. In Proc. WMVC 2002
Video  top-lit. Video side-lit.

· Estimation of Multiple Directional Light Sources for Synthesis of Mixed Reality Images.
Yang Wang, Dimitris Samaras In Proc. Pacific Graphics 2002, pp. 38-47.

· Estimation of Multiple Illuminants from a Single Image of Arbitrary Known Geometry
Yang Wang, Dimitris Samaras, In Proc. ECCV 2002. pp. III:272-288
Slides from  the ECCV 02 presentation and some results from this paper.

Before 2001

· Variable Albedo Surface Reconstruction from Stereo and Shape from Shading.
Dimitrios Samaras, Dimitris Metaxas, Pascal Fua, Y.G.Leclerc. In Proc. CVPR 2000, pp. I:480-487.
Some results from this paper.

· Coupled Lighting Direction and Shape Estimation from Single Images.
Dimitrios Samaras and Dimitris Metaxas. In Proc. ICCV 1999, pp. 868-874

· Incorporating Illumination Constraints in Deformable Models.
Dimitrios Samaras and Dimitris Metaxas. In Proc CVPR  1998,  pp. 322-329.

More to come...