PERSONAL INFORMATIONName: Oana Gabriela CulaCitizenship: Romanian. Office Address: Johnson&Johnson CPPW 199 Grandview Road, Skillman, NJ 908 874 2724 office phone oanacula at caip.rutgers.edu www.caip.rutgers.edu/~oanacula Current Status: Post-Doctoral Scientist Skin Research Center, Methods and Models Development, Johnson&Johnson CPCUS, Skillman NJ. Research Advisor: Dr. Nikiforos Kollias. GRADUATE RESEARCHGraduate Research Advisor: Professor Kristin J. DanaAcademic Advisor: Professor Casimir A. Kulikowski Research Interests: Computer vision, appearance-based modeling of 3D textures, skin texture, image understanding, medical imaging, computer graphics, machine learning. EDUCATIONRutgers University, 1998-2005PhD, Division of Computer & Information Sciences, October 2005 GPA: 3.97/4.0 M.Sci., Department of Computer Science, May 2001 GPA: 4.0/4.0 Polytechnic University of Bucharest, 1991-1997 M.Sci., Applied Electronics and Information Theory, July 1997 Department of Electronics and Telecommunications Graduating mark: 10.0/10.0 B.S., Applied Electronics and Information Theory, July 1996 Department of Electronics and Telecommunications Graduating mark: 9.83/10.0 PROFESSIONAL EXPERIENCEPost-Doctoral Scientist (12/2006 - present)Skin Research Center, Methods and Models Development, Johnson&Johnson CPCUS, Skillman NJ. Research Advisor: Dr. Nikiforos Kollias. Member of Program Committee, IEEE Conference on Computer Vision and Pattern Recognition 2006. Reviewer for Optical Society of America - Applied Optics Journal 2005. Reviewer for IEEE Conference on Computer Vision and Pattern Recognition 2005. Reviewer for EURASIP Journal on Applied Signal Processing 2005. Reviewer for IEEE Transactions on Pattern Analysis and Machine Intelligence 2004. Reviewer for Eurographics Symposium on Rendering 2004. Reviewer for IEEE Transactions on Image Processing 2003. Reviewer for IEEE Conference on Computer Vision and Pattern Recognition 2003. Reviewer for Image and Vision Computing Journal 2002. Reviewer for Pattern Analysis and Applications Journal 2002. Reviewer for Eurographics Workshop on Rendering 2002. Reviewer for IEEE Conference on Computer Vision and Pattern Recognition 2001. Graduate Research Assistant (09/2000 - 08/2005) Computer Vision Laboratory, CAIP Center, Rutgers University. Research advisor: Dr. Kristin J. Dana. Summer Intern (05/2000 - 09/2000) Siemens Corporate Research, Princeton, NJ. Supervisor: Dr. Arun Krishnan. Job description: Integrated visualization and segmentation using volume rendering; creation of polygonal models from binary segmented volumes. Graduate Teaching Assistant (08/1998 - 08/2000) Division of Computer & Information Sciences, Rutgers University. Java instructor for 18 months, also taught courses in Numerical Analysis and Design and Analysis of Computer Algorithms. Teaching Assistant and Research Collaborator (10/1997 - 08/1998) Department of Electronics and Telecommunications, Polytechnic University of Bucharest, Romania. Taught courses in Information Transmission Theory, and Speech Signal Processing. Research Advisor: Dr. Inge Gavat. Research focused on development of speech recognition methods based on hybrid connectionist-statistical approaches. Research Assistant (07/1996 - 09/1997) Department of Electronics and Telecommunications, Polytechnic University of Bucharest, Romania. Research focused on recognition methods for speech and wavelet-based pitch estimation (with Dr. Inge Gavat) and syntactic self-organizing neural approach for automatic Romanian-English translation (with Dr. Victor E. Neagoe). SELECTED PUBLICATIONSJournal Articles (refereed)"3D Texture Recognition Using Bidirectional Feature Histograms", O.G. Cula and K.J. Dana, International Journal of Computer Vision, Vol. 59(1): 33-60, August 2004 (submitted April 2002, submitted revised May 2003, submitted camera-ready September 2003). "Skin Texture Modeling", O.G. Cula, K.J. Dana, F.P.Murphy, and B.K. Rao International Journal of Computer Vision, Vol. 62: No. 1-2, pp. 97-119, April-May 2005 (submitted November 2002, submitted revised April 2003, submitted camera-ready June 2003). "Bidirectional Imaging and Modeling of Skin Texture", O.G. Cula, K.J. Dana, F.P.Murphy, and B.K. Rao IEEE Transactions on Biomedical Engineering, Vol. 51: No. 12, pp. 2148-2159, December 2004 (submitted September 2003, submitted revised February 2004, submitted camera-ready May 2004). "Statistical and Hybrid Methods for Speech Recognition in Romanian", Z. Valsan, I. Gavat, B. Sabac, O. G. Cula, O. Grigore, D. Militaru and O. Dumitru, International Journal of Speech Technology, Vol. 5, No. 3, pp. 259-268, September 2002, Kluwer Academic Publishers. Conference Proceedings (refereed) "Polarization Multiplexing for Bidirectional Imaging", O.G. Cula, K.J. Dana, D. K. Pai, and D. Wang, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2005, vol. II, pp. 1116-1123, San Diego, Ca, June 2005. Acceptance rate: 20%. "Bidirectional Imaging and Modeling of Skin Texture", O.G. Cula, K.J. Dana, F.P. Murphy, and B.K. Rao, Proceedings of Texture 2003 - The 3rd international workshop on texture analysis and synthesis, pp. 13-18, October 17, 2003, Nice, France (in conjunction with International Conference on Computer Vision ICCV 2003). "Image-based Skin Analysis", O.G. Cula and K.J. Dana, Proceedings of Texture 2002 - The 2nd international workshop on texture analysis and synthesis, pp. 35-41, June 1st, 2002, Copenhagen, Denmark (in conjunction with European Conference on Computer Vision ECCV 2002). "Compact Representation of Bidirectional Texture Functions", O.G. Cula and K.J. Dana, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, vol. I, pp. 1041-1047, Hawaii, December 2001. Acceptance rate: 30%. "Recognition Methods for 3D Textured Surfaces" , O.G. Cula and K.J. Dana, Proceedings of SPIE Conference on Human Vision and Electronic Imaging VI, 4299, pp. 209-220, San Jose, California, January 2001. "Discriminative Training in Hybrid HMM-MLP Speech Recognition System For Romanian Language", I. Gavat, M. Zirra, O. G. Cula, Proceedings of COST Workshop, pp. 45-48, Rhodes, Greece, September 1997. "Intonation Estimation For Romanian Language", I. Gavat, M. Zirra, O. G. Cula, Proceedings of ESCA Workshop, pp. 149-152, Athens, Greece, September 1997. "Fuzzy Variants of Hard Classification Rules for speech Pattern Recognition", I. Gavat, O. Grigore, M. Zirra, O. G. Cula, Proceedings of NAFIPS'97, pp. 172-178, New York, USA, September 1997. "A Fuzzy Connectionist Approach To Vowel recognition", V.E. Neagoe, O.G. Cula, Proceedings of Romanian-German Workshop, Bucharest, Romania, November 1997. "Hybrid Speech Recognition System with Discriminative Training Applied for Romanian Language", I. Gavat, M. Zirra, O. G. Cula, Proceedings of MELECON'98, Tel-Aviv, Israel, May 1998. Medical Conventions (refereed) "Standardized Digital Photography: A Novel Approach for Directional Imaging", F. P. Murphy, MD, B. K. Rao, MD, O.G. Cula and K.J. Dana, Poster, 61st Annual Meeting of American Academy of Dermatology, March 21-26, 2003, San Francisco, California. Book Chapters "Elements of Synthesis and Recognition of Speech ", I. Gavat, M. Zirra, O. Grigore, Z. Vaslan, B. Sabac, O. Cula, A. Pascu, Chapter 8: "Hybrid Methods for Speech Recognition", pp. 165-190, Bucharest-Printech 2000, Romania, in Romanian. ELECTRONIC DATABASESRutgers Skin Texture Database: Database of detailed skin texture, consisting of two components: (I) normal skin surface appearance for computer vision and graphics, and (II) skin conditions and disorders for dermatology. Computational characterization of skin texture has potential uses in many applications including robust face models for computer vision, realistic rendering for computer graphics, computer-assisted diagnosis for dermatology, topical drug efficacy testing for the pharmaceutical industry and quantitative comparison of consumer products for cosmetic industry. The database is made publicly available for further research: www.caip.rutgers.edu/rutgers_texture (refereed as part of the second journal article mentioned above).INVITED TALKS"Skin Texture Modeling", Johnson&Johnson, Skillman, NJ, Host: Dr. Nikiforos Kollias, February 8th, 2005."Skin texture Modeling and Polarization Multiplexing", CAVE Laboratory, Columbia University, Host: Ko Nishino, December 3rd, 2004. "Bidirectional Imaging and Modeling of Skin Texture", 2003 IBM Graphics and Visualization Student Symposium, IBM T. J. Watson Research Center, Host: Dr. Jim Klosowski, November 10th, 2003. "Skin Texture Modeling", CAIP Industrial Advisory Board Meeting, Host: Dr. Lawrence Rabiner, Rutgers University, September 8th, 2004. "Surface Science for Computer Vision and Graphics", Spring 2005 Student Research Day, Department of Electrical and Computer Engineering, Rutgers University, March 23rd, 2005. "Polarization Multiplexing for Bidirectional Imaging", Spring 2005 Student Research Day, Computer Science Department, Rutgers University, March 21st, 2005. PROGRAMMING SKILLSC/C++, Java, Matlab, Prolog, Lisp, ML, Pascal, Unix shell programming, Splus.GRADUATE COURSE WORK at Rutgers University (chronological order)Foundations of Computer SciencePattern Recognition: Theory and Applications Design and Analysis of Data Structures and Algorithms Linear Programming Programming Languages and Compilers Parallel Computation: Algorithms and Complexity Machine Learning Image Understanding Computational Geometry Topics in Computers in Biomedicine Robust Computer Vision (Department of Electrical and Computer Engineering) Interpretation of Data (Department of Statistics). SELECTED COURSE-RELATED PROJECTS at Rutgers UniversityCompiler for Context-Free GrammarsA yacc-like tool, developed using Prolog, which produces a parser for an arbitrary context free grammar. Object Recognition System Design and implementation of an object recognition system consisting of acquisition of coupled range/color images, extraction and grouping of surface and shape features, graph representation and graph matching for best object region correspondences, object identity and space location. Protein Structure Characterization by Distance Feature Vectors. Correlation with Secondary Structure Starting from protein domains with known three-dimensional structure, we derive descriptive feature vectors based on distance matrices for locally characterizing the structural configuration. From the distribution of the feature vectors, by grouping them according to some similarity criterion in the feature space, we generate a new vocabulary of representatives. The labeling scheme induced by this vocabulary yields a novel description of the local structure of proteins. We correlate our novel approach with the corresponding secondary structure characterization. LANGUAGES SPOKENEnglish (fluent)French (proficient) Romanian (native) REFERENCESAvailable upon request. |
