Ask Your Question

Haar Like and Adaboost

asked 2017-02-07 10:00:59 -0500

Hello everyone; in documentation of opencv for training haar like there is a Boosted classifer parameters. is haar like trained with adaboost ? those methodes are combined?? i don't get it ! Thanks in advance

edit retag flag offensive close merge delete

1 answer

Sort by » oldest newest most voted

answered 2018-03-06 03:51:59 -0500

Joshitha gravatar image

Yes, the AdaBoosting algorithm is used for Haar based Classifier training.

Essentially we combine weak classifiers into our classifier cascade to hopefully form a strong learner, by way of boosting.

AdaBoosting algorithm (aka Adaptive Boosting)

This algorithm was initially proposed by Yoav Freund and Robert Schapire in A Decision-Theoretic Generalization of on-Line Learning and an Application to Boosting. It is used to improve a learning algorithms performance, by calling weak classifiers repeatedly and updating weights.

Hence you will find these boosted classifer parameters to choose from for your opencv_traincascade utility:

-bt <{DAB, RAB, LB, GAB(default)}>

DAB - Discrete AdaBoost, RAB - Real AdaBoost, LB - LogitBoost, GAB - Gentle AdaBoost.

edit flag offensive delete link more

Question Tools

1 follower


Asked: 2017-02-07 10:00:59 -0500

Seen: 277 times

Last updated: Mar 06 '18