0

I'm working with openCV and I'm a newbie in this field. I'm researching about Camshift. I want to extend this method by using multiple histograms. It means when tracking an object has many than one apperance (ex: rubik cube with six apperance), if we use only one histogram, Camshift will most likely fail.

I know calcHist function in openCV (http://docs.opencv.org/modules/imgproc/doc/histograms.html#calchist) has a parameter is "accumulate", but I don't know how to use and when to use (apply for camshiftdemo.cpp in opencv samples folder). This function can help me solve this problem? Or I have to use difference solution?

I have an idea, that is: create an array histogram for object, for every appearance condition that strongly varies in color, we pre-compute and store all to this array. But when we compute new histogram? It means that the pre-condition to start compute new histogram is what?

And what happend if I have to track multiple object has same color?

Everybody please help me. Thank you so much!

gaara_vn
  • 1
  • Use multiple camshifts on a single target and average their results. For objects of the same color during a "collision" (occlusion), use motion analysis to predict the positions. To detect occlusions, always check the overlap of the boxes (using overloaded operator `&`). – LovaBill Jun 17 '14 at 07:45
  • Tks for comment, William. But I dont know this sentence "using overloaded operator &". Do you have a demo? And motion analysis you mentioned, example that is Kalman filter, right? I hope you can help me. Thank you so much! – gaara_vn Jun 17 '14 at 16:14
  • Deary friend, to track Rubik's cube you need to use grayscale description cues and not the color. Grayscale descriptors are appropriate for your task. Have you tried the "Compressive Tracker"? Google it. It uses Haar-like features. Also, a multi-target tracker has its own challenges as well. But, the very first thing you do is [this](http://stackoverflow.com/questions/8714101/how-can-one-easily-detect-whether-2-rois-intersects-in-opencv). – LovaBill Jun 18 '14 at 07:41

0 Answers0