Search found 157 matches

by Creator
Mon Oct 16, 2017, 19:34
Forum: Direct Graphical Models
Topic: DGMLib Parameter Estimation - Questions
Replies: 3
Views: 9370

Re: DGMLib Parameter Estimation - Questions

I am very glad that the DGM library helped in you research so much. Thank you for citing it in your publication. I would be happy having a chance to read it.

According to the Powell method, I think about a simple changing the sign of the objective function to find its minimum instead of maximum.
by Creator
Thu Sep 28, 2017, 20:36
Forum: Direct Graphical Models
Topic: Regarding DGM installtion with cmake on Linux
Replies: 2
Views: 8401

Re: Regarding DGM installtion with cmake on Linux

Currently Linux is not supported by the DGM library.
However, if you manage to build it in Linux successfully, please share your CMake files and other configurations.
by Creator
Sun Sep 24, 2017, 17:33
Forum: Direct Graphical Models
Topic: Training of a random model.
Replies: 10
Views: 50313

OpenCV Artificial Neural Network Model

Artificial Neural Network (ANN) classifier is inspired by the biological neural networks that constitute mammal brains. The original goal of such approaches was to solve problems in the same way that a human brain would. Hence, ANN classifiers learn (or progressively improve performance) by consider...
by Creator
Sun Sep 24, 2017, 17:26
Forum: Direct Graphical Models
Topic: A few questions about DGM directed and undirected edges
Replies: 4
Views: 10786

Re: A few questions about DGM directed and undirected edges

The theme of the graph transformation is very interesting. The CGraphExt class is to derive from CGraphLayered class the operations of the marginalization, conditioning (partially implemented) and grouping on graph nodes. This operators are based on these two papers: ⋅  Ancestral Graphs an...
by Creator
Wed Sep 20, 2017, 22:49
Forum: Direct Graphical Models
Topic: A few questions about DGM directed and undirected edges
Replies: 4
Views: 10786

Re: A few questions about DGM directed and undirected edges

Thank you for interest to the DGM library. Here are the answers to your questions: ⋅  The arc is the undirected edge in CRFs models. The DGM library emulates every undirected arc with two opposite-directed directed edges. Thus A -- B is represented with A -> B and A <- B. In the exact deco...
by Creator
Thu Aug 24, 2017, 00:21
Forum: Direct Graphical Models
Topic: DGM Visual studio 2017
Replies: 1
Views: 6883

Re: DGM Visual studio 2017

The latest version of DGM library (1.5.3 beta) was built with the Visual Studio 2017; no errors occurred. If you describe the problem in more details, I will try to help you. (for which exactly function the warning appears? which version of the library do you use?) I would like also to encourage you...
by Creator
Wed Aug 23, 2017, 01:31
Forum: Direct Graphical Models
Topic: Training of a random model.
Replies: 10
Views: 50313

OpenCV k-Nearest Neighbors Model

The same as above, it is also possible to make use of the OpenCV implementation of the GMM. It is based on the Expectation Maximization (EM) method and produces the results, depicted at Fig. 10 (Each class is approximated with 16 Gaussians, default parameters). PDF[ featureVector ][ state ] = CvKNN(...
by Creator
Sat Aug 19, 2017, 00:02
Forum: Direct Graphical Models
Topic: Random model training error
Replies: 6
Views: 13530

Re: Random model training error

yes, the problem lies in your ground truth image. One mistake could appear during the automatic conversion from RGB to grayscale image (desaturation). Here different colors may be converted to the same gray value, e.g. color (255, 0, 0) and color (0, 255, 0) by many desaturation algorithms are conve...
by Creator
Fri Aug 11, 2017, 21:25
Forum: Direct Graphical Models
Topic: Random model training error
Replies: 6
Views: 13530

Re: Random model training error

I use mask when the original image needs to choose ROI. For instance, the picture in the attachment has building, road and so on corresponding to white area of the second picture. Hence I don't need to extract features of original image in black and to train black areas. Now I understand the purpos...
by Creator
Sun Jul 30, 2017, 21:43
Forum: Direct Graphical Models
Topic: Random model training error
Replies: 6
Views: 13530

Re: Random model training error

Hi Bryan, here are the answers to your questions: ⋅  The ordinary image (I assume, it is a 3-channel RGB image) will be considered as 3 distinct features. The DGM algorithms will work with that data, but there are some issues: ⋅  Using different features will lead to slightly dif...

Go to advanced search