CvHMM Icon


Discrete Hidden Markov Models based on OpenCV

4.0 Stars (6)
9 Downloads (This Week)
Last Update:
Browse All Files


This project (CvHMM) is an implementation of discrete Hidden Markov Models (HMM) based on OpenCV. It is simple to understand and simple to use. The Zip file contains one header for the implementation and one main.cpp file for a demonstration of how it works. Hope it becomes useful for your projects.

CvHMM Web Site


  • The Viterbi implementation is based on Viterbi Algorithm in Wikipedia.
  • The Baum-Welch Training method is implemented based on "a revealing introduction to hidden markov models"
  • The inputs and outputs of CvHMM class is handled by cv::Mat (OpenCV)
  • The syntax of the member functions is as simple as HMM in MATLAB.


Other Useful Business Software

WhatsUp® Gold - Start A Free 30-Day Trial Icon

All-in-one monitoring of your entire infrastructure with the industry's most user friendly pricing. Free trial of our award-winning software

WhatsUp® Gold - Start A Free 30-Day Trial Icon
If you are like the rest of our user community, your IT team is busy. With pressure to deliver on-time projects, you don’t have a lot of time to spend making your management tools work. You need network monitoring tools that work for you. You want tools that makes it easy to find performance issues before your users do and resolve them before they impact the business. That’s why tens of thousands of customers around the world love WhatsUp Gold.

User Ratings

ease 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5
features 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5
design 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5
support 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5
Write a Review

User Reviews

  • 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5

    Hi, questions: According to your sample_output: INIT: 1 0 0. But many (most) generated states do not start with first state. State 0 should be starting state, thus be always the first of the generated states. Moreover: the type of model is defined through the TRANS data (e.g., left-right ...). If an entry is 0 it should not change during training, otherwise a left-right might become a right-left (that is not clearly not desireable). Also, the log probabilities are very low, thus all sequences have a chance close to 0 to be generated by the model. Comments?

    Posted 11/14/2014
  • 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5

    Nice, thank you

    Posted 06/05/2013
Read more reviews

Additional Project Details

Intended Audience

Information Technology, Science/Research, Education, Telecommunications Industry, Engineering

Programming Language




Thanks for helping keep SourceForge clean.

Screenshot instructions:
Red Hat Linux   Ubuntu

Click URL instructions:
Right-click on ad, choose "Copy Link", then paste here →
(This may not be possible with some types of ads)

More information about our ad policies

Briefly describe the problem (required):

Upload screenshot of ad (required):
Select a file, or drag & drop file here.

Please provide the ad click URL, if possible:

Get latest updates about Open Source Projects, Conferences and News.

No, thanks