Search found 157 matches
- Mon May 08, 2017, 22:07
- Forum: Direct Graphical Models
- Topic: Classification regularization using CRF
- Replies: 5
- Views: 14078
Re: Classification regularization using CRF
As far as I understand, you can use the DGM library for your code. All the examples for the DGM library are given for 2D cases, but it is not difficult to extend the model for a 3D case. First of all, you have to build a graphical model, using the CGraph class (STAGE 1). The graphical model has node...
- Wed May 03, 2017, 21:26
- Forum: Direct Graphical Models
- Topic: Classification regularization using CRF
- Replies: 5
- Views: 14078
Re: Classification regularization using CRF
Thanks for the interest to the DGM library. Have you already tried out the tutorials? This one should help you for a simple classification / segmentation http://research.project-10.de/dgm/doc/a00008.html The code there is well commented and explained. And it is very similar to the code, I used for t...
- Mon Apr 24, 2017, 21:38
- Forum: Direct Graphical Models
- Topic: DGM lib. demo 'Train' question
- Replies: 1
- Views: 6979
Re: DGM lib. demo 'Train' question
Actually the same graph structure from Demo Train may be used for both CRF and MRF. The only difference between CRF and MRF is how you train your model, i.e. in the data for your potential functions (feature vectors). For MRF, the node feature vectors are calculated from a corresponding image pixel,...
- Wed Apr 19, 2017, 20:23
- Forum: Direct Graphical Models
- Topic: Using DGM library with CMake
- Replies: 1
- Views: 23366
Using DGM library with CMake
The easiest way to include the DGM library into your project is to use the find_package( DGM ) command in your CMakeLists.txt file, e.g. : find_package(DGM 1.5 REQUIRED PATHS "$ENV{DGMDIR}/build/install") where "$ENV{DGMDIR}/build/install" is the path to the directory, containing...
- Fri Apr 14, 2017, 19:44
- Forum: Direct Graphical Models
- Topic: Mixed Graphical Models with DGM
- Replies: 3
- Views: 10350
Re: Mixed Graphical Models with DGM
Hello 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 exac...
- Thu Sep 08, 2016, 03:23
- Forum: Direct Graphical Models
- Topic: Sparse Coding Features Extraction
- Replies: 0
- Views: 12983
- Tue Jan 26, 2016, 02:03
- Forum: Direct Graphical Models
- Topic: Parallel computing with DGM ?
- Replies: 1
- Views: 6174
Re: Parallel computing with DGM ?
In order to make use of parallel computing, please uncomment
in the types.h file in the include\ directory.
After that, rebuild the library.
Code: Select all
#define USE_PPL
in the types.h file in the include\ directory.
After that, rebuild the library.
- Tue Jan 26, 2016, 01:43
- Forum: Direct Graphical Models
- Topic: training with multiple images
- Replies: 2
- Views: 6646
Re: training with multiple images
Yes, the training is cumulative: you can train different images in a loop. Functions addFeatureVec() and addFeatureVecs() just gather the train data, so you may use a procedure like: // ========================= STAGE 2: Training ========================= printf("Training... "); ticks = ge...
- Tue Jan 26, 2016, 01:40
- Forum: Direct Graphical Models
- Topic: DGM lib licensing model
- Replies: 2
- Views: 15487
Re: DGM lib licensing model
the current version of the DGM library (1.4.x) is provided under the 3-clause BSD License . So it is friendly to a commercial use. However, the library includes also some methods from the Microsoft Research, e.g. classes CDecodeTRW and CTrainNodeMsRF, which are free for non-commercial use only ( Rea...
- Tue Jan 26, 2016, 01:37
- Forum: Direct Graphical Models
- Topic: Linux instalation of DGM
- Replies: 2
- Views: 9067
Re: Linux instalation of DGM
Hello,
currently the DGM library is designed only for Windows. However, please read this post
currently the DGM library is designed only for Windows. However, please read this post