| Paper: | TA-L4.7 |
| Session: | Image and Video Restoration and Enhancement I |
| Time: | Tuesday, September 18, 12:10 - 12:30 |
| Presentation: |
Lecture
|
| Title: |
IMAGE DENOISING WITH NONPARAMETRIC HIDDEN MARKOV TREES |
| Authors: |
Jyri Kivinen; ICSI, University of California, Berkeley | | |
| | Erik Sudderth; University of California, Berkeley | | |
| | Michael Jordan; University of California, Berkeley | | |
| Abstract: |
We develop a hierarchical, nonparametric statistical model for wavelet representations of natural images. Extending previous work on Gaussian scale mixtures, wavelet coefficients are marginally distributed according to infinite, Dirichlet process mixtures. A hidden Markov tree is then used to couple the mixture assignments at neighboring nodes. Via a Monte Carlo learning algorithm, the resulting hierarchical Dirichlet process hidden Markov tree (HDP-HMT) model automatically adapts to the complexity of different images and wavelet bases. Image denoising results demonstrate the effectiveness of this learning process. |
