|
|
| |
|
|
|
|
 |
Erik Sudderth
|
 |
UC Berkeley, EECS Department
527 Soda Hall #1776
Berkeley, CA 94720-1776
Tel: (510) 642-9582
|
I am a postdoctoral scholar at UC Berkeley, where I work with Professors Michael Jordan and Stuart Russell. My research explores computer vision systems which detect, recognize, and track objects in complex natural scenes. I develop and apply a variety of statistical tools, including graphical models and nonparametric Bayesian methods.
Prior to my arrival at Berkeley, I completed my Ph.D. in
the Electrical Engineering and Computer Science Department at MIT,
working with Professors Alan
Willsky and William Freeman.
|
|
Research &
Publications
|
|
|
|
|
Projects:
Theses:
- Graphical Models for Visual Object Recognition and Tracking.
- Doctoral Thesis, Massachusetts Institute of Technology, May 2006.
- Embedded Trees: Estimation of
Gaussian Processes on Graph with Cycles.
-
Masters Thesis, Massachusetts Institute of Technology, Feb. 2002.
Papers:
- Shared Segmentation of Natural Scenes Using Dependent Pitman-Yor Processes.
- E. Sudderth and M. Jordan.
To appear at Neural Information Processing Systems, Dec. 2008.
- Nonparametric Bayesian Learning of Switching Linear Dynamical Systems.
- E. Fox, E. Sudderth, M. Jordan, and A. Willsky.
To appear at Neural Information Processing Systems, Dec. 2008.
- An HDP-HMM for Systems with State Persistence.
- E. Fox, E. Sudderth, M. Jordan, and A. Willsky.
International Conference on Machine Learning, July 2008.
- Describing Visual Scenes Using Transformed Objects and Parts.
- E. Sudderth, A. Torralba, W. Freeman, and A. Willsky.
International Journal of Computer Vision 77, May 2008.
- Signal and Image Processing with Belief Propagation.
- E. Sudderth and W. Freeman.
DSP Application Column, IEEE Signal Processing Magazine, Mar. 2008.
- Loop Series and Bethe Variational Bounds in Attractive Graphical Models.
- E. Sudderth, M. Wainwright, and A. Willsky.
Neural Information Processing Systems, Dec. 2007.
- Learning Multiscale Representations of Natural Scenes Using Dirichlet Processes.
- J. Kivinen, E. Sudderth, and M. Jordan.
IEEE International Conference on Computer Vision, Oct. 2007.
- Image Denoising with Nonparametric Hidden Markov Trees.
- J. Kivinen, E. Sudderth, and M. Jordan.
IEEE International Conference on Image Processing, Sep. 2007.
- Hierarchical Dirichlet Processes for Tracking Maneuvering Targets.
- E. Fox, E. Sudderth, and A. Willsky.
To appear at the International Conference on Information Fusion, July 2007.
- Depth from Familiar Objects: A Hierarchical Model for 3D Scenes.
-
E. Sudderth, A. Torralba, W. Freeman, and A. Willsky.
IEEE Conference on Computer Vision & Pattern Recognition, June 2006.
- Describing Visual Scenes using Transformed Dirichlet Processes.
-
E. Sudderth, A. Torralba, W. Freeman, and A. Willsky.
Neural Information Processing Systems, Dec. 2005.
- Learning Hierarchical Models of Scenes, Objects, and Parts.
-
E. Sudderth, A. Torralba, W. Freeman, and A. Willsky.
International Conference on Computer Vision, Oct. 2005.
- Distributed Occlusion Reasoning for
Tracking with Nonparametric Belief Propagation.
-
E. Sudderth, M. Mandel, W. Freeman, and A. Willsky.
Neural Information Processing Systems, Dec. 2004.
- Embedded Trees: Estimation of Gaussian
Processes on Graphs with Cycles.
-
E. Sudderth, M. Wainwright, and A. Willsky.
IEEE Transactions on Signal Processing 52(11), Nov. 2004.
An earlier version appeared as MIT LIDS Technical
Report 2562, Apr. 2003.
- Visual Hand Tracking Using Nonparametric
Belief Propagation.
-
E. Sudderth, M. Mandel, W. Freeman, and A. Willsky.
Workshop on Generative Model Based Vision, CVPR, June 2004.
- Efficient Multiscale Sampling from Products
of Gaussian Mixtures.
-
A. Ihler, E. Sudderth, W. Freeman, and A. Willsky.
Neural Information Processing Systems, Dec. 2003.
- Nonparametric Belief Propagation.
-
E. Sudderth, A. Ihler, W. Freeman, and A. Willsky.
IEEE Conference on Computer Vision & Pattern Recognition, June 2003.
An earlier version appeared as MIT LIDS Technical
Report 2551, Oct. 2002.
- Projection Algebra Analysis of Error-Correcting
Codes.
-
J. Yedidia, E. Sudderth, and J-P. Bouchaud.
Allerton Conference on Communication, Control, and Computing, Oct. 2001.
- Tree-Based Modeling and Estimation of
Gaussian Processes on Graphs with Cycles.
-
M. Wainwright, E. Sudderth, and A. Willsky.
Neural Information Processing Systems, Dec. 2000.
|
| |
Software
|
|
|
|
|
Transformed Dirichlet Process (TDP) Matlab Toolbox
Implements learning algorithms for the models in our 2008 IJCV paper.
Software Only (208 KB .zip)
Software, Demos, and Image Databases (355 MB .zip)
Nonparametric Belief Propagation
Implements efficient multiscale sampling algorithms from our 2003 NIPS paper.
Kernel Density Estimation Matlab Toolbox (written by Alex Ihler and Mike Mandel) |
|
|
|
|
Teaching |
|
|
|
|
|
|
 |
|
| |
|
|
