Ask Your Question

Can I "retrain" OpenCV detection .xml model which is ready?

asked 2016-03-25 06:01:25 -0500

Pea gravatar image

About the ready detection model that offered by OpenCV, like lbpcascade_frontalface.xml and haarcascade_frontalface_default.xml, etc.

I know I can use myself databases to train the new model.

but... If I want to use the ready .xml model and "additional add" my own databases to "retrain" the ready detection model. Is it OK to do? And how can I do?

I want to detect frontal face and hope to use the ready lbpcascade_frontalface.xml and add other training samples to improve the model. I want to add the negative samples to retrain and reduce the false positive rate and increse the detect speed.

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2016-03-25 06:09:58 -0500

Then you need both the original data and your new data. Add your new detections as hard positives at the beginning of your positive training set to enforce them being used and to ensure that they add more knowledge to the model.

edit flag offensive delete link more


Take a look at bootstrapping object models, which is exactly how this is called!

StevenPuttemans gravatar imageStevenPuttemans ( 2016-03-25 06:10:19 -0500 )edit

Also, as I might add, it could be interesting to take a look at the OpenCV 3 Blueprints book, which discusses optimal object detection and the principle of bootstrapping.

StevenPuttemans gravatar imageStevenPuttemans ( 2016-03-29 06:10:55 -0500 )edit

Question Tools

1 follower


Asked: 2016-03-25 06:01:25 -0500

Seen: 217 times

Last updated: Mar 25 '16