I just got Tbb running on open cv using gcc and g++ . i did disable the -Werror flag in the makefile.tbb and now the facedetect code is effectively parallelized on my quad core ... though far from optimized (speedup 2.8x to 3.2x). i get 3.6x on open mp parallelization I did run into another problem though ..... i want to read 2 cameras in parallel with tbb.
the camera read routine is an infinite loop like
it compiles all well but only reads one loop at a time even though 2 of these loops should be running parallelly.
once i manually destroy the window of one thread the other thread starts....
if i make it as a read of 2 images (disabling the loop) it works fine for one frame each... so i implemented the infinite loop outside the parallel for call
which led to immense slowdown.
less than a frame per second....
thread checker showed me that a thread has been waiting for more than three seconds....