Take the 2-minute tour ×
Signal Processing Stack Exchange is a question and answer site for practitioners of the art and science of signal, image and video processing. It's 100% free, no registration required.

I am working on real time video compression. I am modeling it with MATLAB. I need to later implement it on DSP processor with 5832MIPS, 729MHz.

Is it feasible to implement Gaussian Mixture models on DSP processor or are there better algorithms for object detection ?

Thanks in advance

share|improve this question
1  
Tell us more about your algorithm (reference to a paper ?) - GMM is a very generic technique which can be used in many different ways, and in many different contexts. Will the training run on your embedded system too, or is it just a matter of running the classification/evaluation code? Will there be some online training/adaptation? What are you going to classify? Pixels? Regions? A single feature vector extracted on the whole image? –  pichenettes Dec 10 '12 at 14:00
    
Reference paper that i'm using is 'GMM-based efficient foreground detection with adaptive region update',ICIP,Hao Li. –  user1752 Dec 11 '12 at 1:43
    
The training need not run on the embedded system. I'm going to classify regions. –  user1752 Dec 11 '12 at 1:50
    
As mentioned by others your question hard to answer without knowing more about the context. The quality of your classifier will of course depend on the model type used, but often feature extraction, and other pre-processing steps are the actual part leading to either failure or success of your product. –  KlausCPH Jan 18 '13 at 20:28
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.