Description: The EM algorithm is short for Expectation-Maximization algorithm. It is based on an iterative optimization of the centers and widths of the kernels. The aim is to optimize the likelihood that the given data points are generated by a mixture of Gaussians. The numbers next to the Gaussians give the relative importance (amplitude) of each component.-The EM algorithm is short for Expectation - Maximization algorithm. It is based on an ITERA tive optimization of the centers and widths of t he kernels. The aim is to optimize the likelihoo d that the given data points are generated by a mi xture of Gaussians. The numbers next to the Gaus sians give the relative importance (amplitude ) of each component. Platform: |
Size: 15614 |
Author:陈伟 |
Hits:
Description: The EM algorithm is short for Expectation-Maximization algorithm. It is based on an iterative optimization of the centers and widths of the kernels. The aim is to optimize the likelihood that the given data points are generated by a mixture of Gaussians. The numbers next to the Gaussians give the relative importance (amplitude) of each component.-The EM algorithm is short for Expectation- Maximization algorithm. It is based on an ITERA tive optimization of the centers and widths of t he kernels. The aim is to optimize the likelihoo d that the given data points are generated by a mi xture of Gaussians. The numbers next to the Gaus sians give the relative importance (amplitude ) of each component. Platform: |
Size: 15360 |
Author:陈伟 |
Hits:
Description: 使用混合高斯函数，对点配准的方法，他的鲁棒性比较好。
-This package contains the MATLAB code for the robust point-set
registration algorithm discribed in the A Robust Algorithm for Point Set Registration Using Mixture of Gaussians."
Platform: |
Size: 39936 |
Author:wangwei |
Hits:
Description: KlustaKwik is an open-source C++ program for automatic clustering of continuous data into a mixture of Gaussians. The program was originally developed for sorting of neuronal action potentials, but can be applied to any sort of data. Platform: |
Size: 46080 |
Author:wang |
Hits:
Description: Vehicle Tracking using a background subtraction based on mixture of Gaussians, and Kalman filtering to remove noise.
Require OpenCV to be installed.
By Jonathan Gagne
University of Waterloo
jgagne@uwaterloo.ca Platform: |
Size: 13237248 |
Author:jon |
Hits:
Description: Motion Tracking
=== === ===
This tarball contains all code required to run the tracking algorithm
on a sequence of images. Run the file run_tracker.m in Matlab and
follow the instructions. You will need to have a directory of
sequentially numbered images available. After entering the path and
file types the tracker will begin processing. Once the data window
appears the algorithm begins building a background model and attempts
to track objects. By clicking on any of the four subwindows you can
investigate the background representation (a Mixture of Gaussians) of
any pixel. The two windows that then appear display the mixture once
as a two-dimensional scatter plot (ignoring the blue colour
component), and once as a one-dimensional evolution of the red colour
component only. These plots make the internal processing visible and
should help determining suitable parameters to be set in
mixture_parameters.m.-Motion Tracking
===============
This tarball contains all code required to run the tracking algorithm
on a sequence of images. Run the file run_tracker.m in Matlab and
follow the instructions. You will need to have a directory of
sequentially numbered images available. After entering the path and
file types the tracker will begin processing. Once the data window
appears the algorithm begins building a background model and attempts
to track objects. By clicking on any of the four subwindows you can
investigate the background representation (a Mixture of Gaussians) of
any pixel. The two windows that then appear display the mixture once
as a two-dimensional scatter plot (ignoring the blue colour
component), and once as a one-dimensional evolution of the red colour
component only. These plots make the internal processing visible and
should help determining suitable parameters to be set in
mixture_parameters.m. Platform: |
Size: 36511744 |
Author:gobsy |
Hits:
Description: Bayesian mixture of Gaussians. This set of files contains functions for performing inference and learning on a Bayesian Gaussian mixture model. Learning is carried out via the variational expectation maximization algorithm. Platform: |
Size: 6144 |
Author:ruso |
Hits:
Description: A common method for real-time segmentation of
moving regions in image sequences involves “background
subtraction,” or thresholding the error between
an estimate of the image without moving objects and
the current image. The numerous approaches to this
problem differ in the type of background model used
and the procedure used to update the model. This paper
discusses modeling each pixel as a mixture of Gaussians
and using an on-line approximation to update
the model. The Gaussian distributions of the adaptive
mixture model are then evaluated to determine which
are most likelyt o result from a background process.
Each pixel is classified based on whether the Gaussian
distribution which represents it most effectivelyis considered
part of the background model. Platform: |
Size: 186368 |
Author:ajinkya |
Hits:
Description: Among the high-complexity methods, two methods dominate the literature Kalman filtering and Mixture of Gaussians (MoG). Both have their advantages, but Kalman filtering gets slammed in every paper for leaving object trails that can t be eliminated. As this seems like a possible deal breaker for many applications, I went with MoG. Also, MoG is more robust, as it can handle multi-modal distributions. For instance, a leaf waving against a blue sky has two modes—leaf and sky. MoG can filter out both. Kalman filters effectively track a single Gaussian, and are therefore unimodal: they can filter out only leaf or sky, but usually not both. -Among the high-complexity methods, two methods dominate the literature Kalman filtering and Mixture of Gaussians (MoG). Both have their advantages, but Kalman filtering gets slammed in every paper for leaving object trails that can t be eliminated. As this seems like a possible deal breaker for many applications, I went with MoG. Also, MoG is more robust, as it can handle multi-modal distributions. For instance, a leaf waving against a blue sky has two modes—leaf and sky. MoG can filter out both. Kalman filters effectively track a single Gaussian, and are therefore unimodal: they can filter out only leaf or sky, but usually not both. Platform: |
Size: 80896 |
Author:mohammed fadhle |
Hits:
Description: 这篇文章是关于如何改进混合高斯法的一个综述，混合高斯法用于目标检测，目标分割。-This article is an overview of how to improve the GMMS ,the GMMS is used for target detection, object segmentation. Platform: |
Size: 1306624 |
Author:黄鹏 |
Hits:
Description: 这个代码适用于检测运动目标，基于混合高斯建模的运动目标检测。-this code can help you detection some objects，such as car，people，and so on. which based on mixture of gaussians background model Platform: |
Size: 2048 |
Author: |
Hits:
Description: C++实现的自动聚类系统KlustaKwik源代码\KlustaKwik-R1-7\KlustaKwik-ks for any type of data. We needed a program that
would:
1) Fit a mixture of Gaussians with unconstrained covariance matrices
2) Automatically choose the number of mixture components
3) Be robust against noise
4) Reduce the problem of local minima
5) Run fast on large data sets (up to 100000 points, 48 dimensions)
Speed in particular was essential. KlustaKwik is based on the CEM algorithm of
Celeux and Govaert (which is faster than the standard EM algorithm Platform: |
Size: 409600 |
Author:大家 |
Hits:
Description: This m-file implements the mixture of Gaussians algorithm for background subtraction.-This m-file implements the mixture of Gaussians algorithm for background subtraction. Platform: |
Size: 2048 |
Author:Nargis |
Hits:
Search - mixture of gaussians