Abstract | ||
---|---|---|
Abstract Many computer,vision problems,can be formulated in a Bayesian framework,with Markov Random Field (MRF) or Conditional Random Field (CRF) priors. Usually, the model assumes that a full Maximum,A Posteriori (MAP) es- timation will be performed for inference, which can be re- ally slow in practice. In this paper, we argue that through appropriate training, a MRF/CRF model can be trained to perform very well on a suboptimal inference algorithm. The model is trained together with a fast inference algorithm through an optimization of a loss function on a training set containing pairs of input images and desired outputs. A validation set can be used in this approach to estimate the generalization performance,of the trained system. We ap- ply the proposed method to an image denoising application, training a Fields of Experts MRF together with a 1-4 iter- ation gradient descent inference algorithm. Experimental validation on unseen data shows that the proposed training approach,obtains an improved benchmark,performance,as well as a 1000-3000 times speedup compared,to the Fields of Experts MRF trained with contrastive divergence. Us- ing the new approach, image denoising can be performed in real-time, at 8fps on a single CPU for a 256 256 image sequence, with close to state-of-the-art accuracy. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/CVPRW.2009.5206811 | CVPR |
Keywords | Field | DocType |
Bayes methods,Markov processes,computer vision,estimation theory,gradient methods,image denoising,image sequences,inference mechanisms,learning (artificial intelligence),random processes,Bayesian framework,MRF/CRF model,Markov random field,computer vision problems,conditional random field,contrastive divergence,image denoising,image sequence,iteration gradient descent inference algorithm,learning real-time MRF inference,maximum a posteriori estimation,suboptimal inference algorithm | Conditional random field,Computer vision,Gradient descent,Pattern recognition,Computer science,Inference,Markov random field,Artificial intelligence,Maximum a posteriori estimation,Prior probability,Bayesian probability,Belief propagation | Conference |
Volume | Issue | ISSN |
2009 | 1 | 1063-6919 |
Citations | PageRank | References |
15 | 0.73 | 24 |
Authors | ||
1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Adrian Barbu | 1 | 768 | 58.59 |