Modeling natural images using gated MRFs
- PMID: 23868780
- DOI: 10.1109/TPAMI.2013.29
Modeling natural images using gated MRFs
Abstract
This paper describes a Markov Random Field for real-valued image modeling that has two sets of latent variables. One set is used to gate the interactions between all pairs of pixels, while the second set determines the mean intensities of each pixel. This is a powerful model with a conditional distribution over the input that is Gaussian, with both mean and covariance determined by the configuration of latent variables, which is unlike previous models that were restricted to using Gaussians with either a fixed mean or a diagonal covariance matrix. Thanks to the increased flexibility, this gated MRF can generate more realistic samples after training on an unconstrained distribution of high-resolution natural images. Furthermore, the latent variables of the model can be inferred efficiently and can be used as very effective descriptors in recognition tasks. Both generation and discrimination drastically improve as layers of binary latent variables are added to the model, yielding a hierarchical model called a Deep Belief Network.
Similar articles
-
Dense photometric stereo: a Markov random field approach.IEEE Trans Pattern Anal Mach Intell. 2006 Nov;28(11):1830-46. doi: 10.1109/TPAMI.2006.224. IEEE Trans Pattern Anal Mach Intell. 2006. PMID: 17063687
-
Posterior-mean super-resolution with a causal Gaussian Markov random field prior.IEEE Trans Image Process. 2012 Jul;21(7):3182-93. doi: 10.1109/TIP.2012.2189578. Epub 2012 Feb 29. IEEE Trans Image Process. 2012. PMID: 22389146
-
Brain tissue classification of magnetic resonance images using partial volume modeling.IEEE Trans Med Imaging. 2000 Dec;19(12):1179-87. doi: 10.1109/42.897810. IEEE Trans Med Imaging. 2000. PMID: 11212366
-
Training an active random field for real-time image denoising.IEEE Trans Image Process. 2009 Nov;18(11):2451-62. doi: 10.1109/TIP.2009.2028254. Epub 2009 Jul 24. IEEE Trans Image Process. 2009. PMID: 19635701
-
Greedy learning of binary latent trees.IEEE Trans Pattern Anal Mach Intell. 2011 Jun;33(6):1087-97. doi: 10.1109/TPAMI.2010.145. IEEE Trans Pattern Anal Mach Intell. 2011. PMID: 20714018
Cited by
-
A Comprehensive Review of High Throughput Phenotyping and Machine Learning for Plant Stress Phenotyping.Phenomics. 2022 Apr 4;2(3):156-183. doi: 10.1007/s43657-022-00048-z. eCollection 2022 Jun. Phenomics. 2022. PMID: 36939773 Free PMC article. Review.
-
Artificial intelligence and its applications in digital hematopathology.Blood Sci. 2022 Jul 14;4(3):136-142. doi: 10.1097/BS9.0000000000000130. eCollection 2022 Jul. Blood Sci. 2022. PMID: 36518598 Free PMC article.
-
Review of the State of the Art of Deep Learning for Plant Diseases: A Broad Analysis and Discussion.Plants (Basel). 2020 Oct 1;9(10):1302. doi: 10.3390/plants9101302. Plants (Basel). 2020. PMID: 33019765 Free PMC article. Review.
-
Cross-modality deep learning-based prediction of TAP binding and naturally processed peptide.Immunogenetics. 2018 Jul;70(7):419-428. doi: 10.1007/s00251-018-1054-6. Epub 2018 Feb 28. Immunogenetics. 2018. PMID: 29492592
-
The visual system's internal model of the world.Proc IEEE Inst Electr Electron Eng. 2015 Aug;103(8):1359-1378. doi: 10.1109/JPROC.2015.2434601. Epub 2015 Jul 6. Proc IEEE Inst Electr Electron Eng. 2015. PMID: 26566294 Free PMC article.
Publication types
LinkOut - more resources
Full Text Sources
Other Literature Sources
Miscellaneous