Ask Your Question

Opencv with both TBB and IPP

asked 2013-05-29 13:14:12 -0600

UserOpenCV gravatar image

I have build Opencv With TBB enabled. And used "detectMultiscale" and wrote a basic program to detect the face. I couldnt find any changes in processing time if there is a face in a frame. Also i noticed that the processing time has been reduced by two times if there is no face in a frame(empty).

1) How to improve the processing time now?

2) Is it worth to go for Intel IPP? What could be the actual benefit?

Can anyone give me an advice?

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2013-05-29 14:35:37 -0600

Probably you used the 2.4.5 library with and without TBB to compare, however, since OpenCV 2.4.3 multithreaded support functionality has been included in the source code, not needing to build openCV with the TBB support anymore. It is done automatically where necessary and the included dll's are contained in the source where needed.

About IPP, don't think that the increase will be worth the payment, since you can only get a 30 day trial version for free.

edit flag offensive delete link more


Thanks for the answer. I used peopledetect.cpp from samples and compiled with both TBB and without TBB in opencv2.4.5. I can see the Opencv2.4.5 compiled with TBB performs 2x speed where as Opencv2.4.5 without TBB performs very slow. And one more help can you conform that by any limks that TBB is enabled by default?

UserOpenCV gravatar imageUserOpenCV ( 2013-05-30 13:08:49 -0600 )edit
StevenPuttemans gravatar imageStevenPuttemans ( 2013-05-31 02:16:55 -0600 )edit

Thanks. But your comment on Hog descriptor? What I mean is hog descriptor performs faster with opencv2.4.5 TBB enabled build. The speed has been incresed two folds compared to opencv2.4.5 without TBB.

UserOpenCV gravatar imageUserOpenCV ( 2013-05-31 02:46:12 -0600 )edit

Question Tools


Asked: 2013-05-29 13:14:12 -0600

Seen: 1,486 times

Last updated: May 29 '13