littlie's profile - activity

You can try several heuristics once you get the features. Some ideas:

1. Discard items that are too long/short. Maybe calculate the mean and standard deviation and discard features that are more than mean + 2standard deviation or less than mean - 2standard deviation.
2. If you have a sequence, tracking vector from N to N+1 should be similar to N-1 to N.
3. Try different feature tracking by OpenCV - FAST, SIFT, SURF etc. Each one has advantages/disadvantages based on the features you are trying to track.

Did you use the vector notation of accessing elements.

Suppose the vector, vec consists of 2 elements and you want to access the second GpuMat:
GpuMat g = vec[1];

I am assuming by members you mean individual element of vector.

Hi,

I am trying to convert Mat to GpuMat so that I can use GPU for processing.

I tried the following two ways:

GpuMat g; 
Mat m; 
1. g = GpuMat(m) -> takes around 2 ms 
2. g.upload(m)   -> takes roughly 1 ms.

I have around 10 Mat to be converted to GpuMat - the time just adds up, is there a quick and efficient way to do the conversion?

Thanks.