Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
On active contour models and balloons
CVGIP: Image Understanding
Knowledge representation and inference in similarity networks and Bayesian multinets
Artificial Intelligence
Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
International Journal of Computer Vision - Special issue on statistical and computational theories of vision: modeling, learning, sampling and computing, Part I
Image Segmentation Based on the Integration of Markov Random Fields and Deformable Models
MICCAI '00 Proceedings of the Third International Conference on Medical Image Computing and Computer-Assisted Intervention
Gradient Vector Flow: A New External Force for Snakes
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Topologically adaptable snakes
ICCV '95 Proceedings of the Fifth International Conference on Computer Vision
Belief Propagation and Revision in Networks with Loops
Belief Propagation and Revision in Networks with Loops
A graphical model framework for coupling MRFs and deformable models
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Factor graphs and the sum-product algorithm
IEEE Transactions on Information Theory
Hi-index | 0.00 |
In this paper we present a new framework for image segmentation using probabilistic multinets. We apply this framework to integration of region-based and contour-based segmentation constraints. A graphical model is constructed to represent the relationship of the observed image pixels, the region labels and the underlying object contour. We then formulate the problem of image segmentation as the one of joint region-contour inference and learning in the graphical model. The joint inference problem is solved approximately in a band area around the estimated contour. Parameters of the model are learned on-line. The fully probabilistic nature of the model allows us to study the utility of different inference methods and schedules. Experimental results show that our new hybrid method outperforms methods that use homogeneous constraints.