Ask Your Question

How can i test the face detection?

asked 2013-08-07 14:27:05 -0500

Adrianos01 gravatar image

updated 2013-08-07 15:21:30 -0500


how can i test the face detection in a Laptop?

the people this question a negative review, "Have their boredom"?

edit retag flag offensive close merge delete



hey, don't be angry at 2 ppl downvoting, but too lazy for telling you the obvious reason:

your question is sloppy and does not show much own research effort.

(before you start bashing me now, none of the downvotes is mine, upvoted the answer instead)

still, apart from rbertc's nice anwer(shouldn't you give him credit for this, btw ?) , you might want to look here for a sample

berak gravatar imageberak ( 2013-08-07 16:14:31 -0500 )edit

1 answer

Sort by ยป oldest newest most voted

answered 2013-08-07 14:38:30 -0500

rbertc gravatar image

What environment are you working in? Opencv comes with some xml classifier files, including a few for face detection. Here is an example of how to use them.

edit flag offensive delete link more


Are u joke me? i know this website and i have a code thats code thats run very well. i mean how can i TEST it. NOT write a code. ;) How SPEED it is or how sure it is. JUST TEST

Adrianos01 gravatar imageAdrianos01 ( 2013-08-07 15:04:13 -0500 )edit

I apologize, I guess I'm having a hard time understanding exactly what you're looking for. Could you try to provide more information, like what "face detection" you're using? Simply saying "how can i test the face detection in a laptop?" is way too vague.

rbertc gravatar imagerbertc ( 2013-08-07 17:16:07 -0500 )edit

You probably mean "evaluate" not test: for me, and I think it's the same for rbertc, test means: compile and execute the code, see if it's working. If you want to evaluate the code, look at the paper on face detection on internet. It exists database of face, with the ROI (rectangle around what is a face and what is not). Therefore, you just have to detect faces on each image, see the overlapping of the ROI and compute a score (#pixels good/#pixels bad, other interesting metrics that can be found on the papers). You also have to evaluate the false positive detection (when you found a face that is something else!). For the speed, it's the same, see how many face you found by second. Be more precise on your question or add some explanation of what you have done, people don't know that!

Mathieu Barnachon gravatar imageMathieu Barnachon ( 2013-08-07 17:49:38 -0500 )edit

@Adrianos01, please change the way you are answering to people and watch your language in your comments. Caps are basically shouting on forums, which is first of all annoying to read and second of all will get your posts deleted. Please put some effort in following the guidelines!

StevenPuttemans gravatar imageStevenPuttemans ( 2013-08-08 02:40:27 -0500 )edit

@StevenPuttermans and @all Does not look like her it is so hard. But you can also ask too: What do you mean by test? ;) Thanks to all.

Adrianos01 gravatar imageAdrianos01 ( 2013-08-10 04:46:48 -0500 )edit

Question Tools


Asked: 2013-08-07 14:27:05 -0500

Seen: 226 times

Last updated: Aug 07 '13