Learning Non-Gaussian Stochastic Systems for Dynamic Textures
نویسندگان
چکیده
Figure 1. A. Factor graph for an LDS with state-space potentials. B. Images corresponding to the training data and outputs of different algorithms. Scatterplots of the first two state component sequences (red = training, blue = simulated), along with the approximate scale. Only our method results in a realistic state simulation and correct clock structure. C. Other videos we are working on: smoke, traffic, flag, hallway.
منابع مشابه
Comparative Study of Random Matrices Capability in Uncertainty Detection of Pier’s Dynamics
Because of random nature of many dependent variables in coastal engineering, treatment of effective parameters is generally associated with uncertainty. Numerical models are often used for dynamic analysis of complex structures, including mechanical systems. Furthermore, deterministic models are not sufficient for exact anticipation of structure’s dynamic response, but probabilistic models...
متن کاملGaussian Process Latent Force Models for Learning and Stochastic Control of Physical Systems
This paper is concerned with estimation and stochastic control in physical systems which contain unknown input signals or forces. These unknown signals are modeled as Gaussian processes (GP) in the sense that GP models are used in machine learning. The resulting latent force models (LFMs) can be seen as hybrid models that contain a first-principles physical model part and a non-parametric GP mo...
متن کاملNovel Radial Basis Function Neural Networks based on Probabilistic Evolutionary and Gaussian Mixture Model for Satellites Optimum Selection
In this study, two novel learning algorithms have been applied on Radial Basis Function Neural Network (RBFNN) to approximate the functions with high non-linear order. The Probabilistic Evolutionary (PE) and Gaussian Mixture Model (GMM) techniques are proposed to significantly minimize the error functions. The main idea is concerning the various strategies to optimize the procedure of Gradient ...
متن کاملMinimum Entropy Parameter Estimation of Bounded Nonlinear Dynamic Systems with Non-Gaussian State and Measurement Noise
Parameter estimation plays an important role in Systems Biology in helping to understand the complex behavior of signal transduction networks. The problem becomes more complex as the inherent stochasticity of the signaling mechanism involves noise components of non-Gaussian nature. In this paper a novel stochastic parameter estimation method has been developed where the entropy of the joint res...
متن کاملIRWIN AND JOAN JACOBS CENTER FOR COMMUNICATION AND INFORMATION TECHNOLOGIES Superresolution of self-similar textures
Texture enhancement presents an ongoing challenge, in spite of the considerable progress made in recent years. Whereas most of the effort has been devoted so far to enhancement of regular textures, stochastic textures, which exhibit fine details, are encountered in most natural images wherein they still present an outstanding problem insofar as superresolution enhancement is concerned. In this ...
متن کامل