Name
Abstract | ||
|---|---|---|
We formulate a model for probability distributions on im- age spaces. We show that any distribution of images can be factored exactly into conditional distributions of feature vectors at one resolution (pyramid level) conditioned on the image information at lower resolutions. We would like to factor this over positions in the pyramid levels to make it tractable, but such factoring may miss long-range depen- dencies. To capture long-range dependencies, we introduce hidden class labels at each pixel in the pyramid. The result is a hierarchical mixture of conditional probabilities, similar to a hidden Markov model on a tree. The model parameters can be found with maximum likelihood estimation using the EM algorithm. We have obtained encouraging preliminary results on the problems of detecting various objects in SAR images and target recognition in optical aerial images. |
Year | DOI | Venue |
|---|---|---|
1999 | 10.1109/ICIP.2000.899375 | NIPS |
Keywords | Field | DocType |
hidden markov models,image resolution,maximum likelihood estimation,object detection,optimisation,probability,radar imaging,synthetic aperture radar,em algorithm,sar images,conditional distributions,conditional probabilities,feature vectors,hidden markov model,hidden class labels,hierarchical image probability models,image distribution,image information,image spaces,long-range dependencies,model parameters,optical aerial images,pixel,probability distributions,pyramid levels,target recognition,training,tree,conditional probability,data analysis,conditional distribution,probability distribution,maximum likelihood estimate,object recognition,feature vector | Computer vision,Feature vector,Conditional probability distribution,Pattern recognition,Conditional probability,Computer science,Expectation–maximization algorithm,Probability distribution,Artificial intelligence,Pyramid,Hidden Markov model,Image resolution | Conference |
Volume | ISSN | ISBN |
3 | 1522-4880 | 0-7803-6297-7 |
Citations | PageRank | References |
6 | 0.76 | 6 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Spence, Clay | 1 | 6 | 0.76 |
| Parra, Lucas | 2 | 6 | 0.76 |
| Sajda, P. | 3 | 6 | 0.76 |