Svetlana Lazebnik

SVETLANA LAZEBNIK

Assistant Professor
Department of Computer Science
University of North Carolina at Chapel Hill
Office: 244 Fred Brooks Building
Phone: (919) 962-1942
Email: lazebnik -at- cs.unc.edu

I received my Ph.D. at the University of Illinois at Urbana-Champaign in May of 2006 under the supervision of Jean Ponce. During the course of my dissertation work, I have also collaborated with Cordelia Schmid and the LEAR group at INRIA Rhône-Alpes. In August of 2007, I have left the cornfields of Central Illinois for the forests of North Carolina. My research specialty is computer vision. The main themes of my research include:

Developing effective image representations for recognition, from low-level patch vocabularies to high-level semantic concepts
Integrating statistical and geometric techniques for comprehensive 2D and 3D scene description
Modeling and organizing large-scale Internet image collections

CURRICULUM VITAE

PUBLICATIONS

TEACHING

Fall 2008: Computational Photography (790-096)
Spring 2008: Computer Vision (776)
Fall 2007: Computer Vision and the Web (790-096)

SELECTED PROJECTS

	Internet Photo Collections Modeling and Recognition of Landmark Image Collections Using Iconic Scene Graphs X. Li, C. Wu, C. Zach, S. Lazebnik and J.-M. Frahm, to appear in ECCV 2008 Project webpage (preliminary) Computing Iconic Summaries for General Visual Concepts R. Raguram and S. Lazebnik, First IEEE Workshop on Internet Vision, 2008
	Quantization-Based Techniques for Supervised and Unsupervised Learning Supervised Learning of Quantizer Codebooks by Information Loss Minimization S. Lazebnik and M. Raginsky, accepted to PAMI AISTATS 2007 slides (PDF) Estimation of Intrinsic Dimensionality Using High-Rate Vector Quantization M. Raginsky and S. Lazebnik, NIPS 2005 NIPS oral presentation slides (PDF)
	Natural Scene Statistics and Global Image Recognition Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories S. Lazebnik, C. Schmid, and J. Ponce, CVPR 2006 Slides (PDF), scene category dataset, Caltech-101 comparison
	Texture and Object Recognition Using Local Features A Discriminative Framework for Texture and Object Recognition Using Local Image Features S. Lazebnik, C. Schmid, and J. Ponce, Toward Category-Level Object Recognition (LNCS 4170), 2006 Birds dataset A Sparse Texture Representation Using Local Affine Regions S. Lazebnik, C. Schmid, and J. Ponce, PAMI 2005 UIUC texture dataset
	3D Object Modeling and Recognition in Images and Video Segmenting, Modeling, and Matching Video Clips Containing Multiple Moving Objects F. Rothganger, S. Lazebnik, C. Schmid, and J. Ponce, PAMI 2007 Video results page 3D Object Modeling and Recognition Using Local Affine-Invariant Image Descriptors and Multi-View Spatial Constraints F. Rothganger, S. Lazebnik, C. Schmid, and J. Ponce, IJCV 2006 3D object database, software, comparative evaluation
	Projective Surface Reconstruction Based on Outlines Projective Visual Hulls S. Lazebnik, Y. Furukawa, and J. Ponce, IJCV 2007 Visual hull datasets The Local Projective Shape of Smooth Surfaces and Their Outlines S. Lazebnik and J. Ponce, IJCV 2005

THESES

Local, Semi-Local and Global Models for Texture, Object and Scene Recognition
Ph.D. Dissertation, May 2006
Defense slides (PowerPoint)
Projective Visual Hulls
M.S. Thesis, December 2002 --- winner of the David J. Kuck Best Master's Thesis Award

PERSONAL

I was born in Kiev, Ukraine, on March 7, 1979. If you want to learn more about this momentous date, you can read this essay I wrote in high school.
My husband Maxim Raginsky and my sister Mariya Lazebnik are brilliant electrical engineers.
I take photos from time to time. Check out my photo album and Flickr account.
I also love collecting interesting images of all sorts. This scrapbook is just the tip of the iceberg of my collection.

Last updated August 18, 2008

SVETLANA LAZEBNIK

CURRICULUM VITAE

PUBLICATIONS

TEACHING

SELECTED PROJECTS

Internet Photo Collections

Quantization-Based Techniques for Supervised and Unsupervised Learning

Natural Scene Statistics and Global Image Recognition

Texture and Object Recognition Using Local Features

3D Object Modeling and Recognition in Images and Video

Projective Surface Reconstruction Based on Outlines

THESES

PERSONAL