Face Detection with Asymmetric Boosting

Asymmetric boosting, while acknowledged to be
important to state-of-the-art face detection, is
typically based on the trial-and-error practice,
rather than on principled methods. This work solves a
number of issues related to asymmetric boosting and
the use of asymmetric boosting in face detection. It
shows how a proper understanding and use of
asymmetric boosting leads to significant improvements
in the
learning time, the learning capacity, the detection
speed and the detection accuracy of a face detector.

There are four main contributions in this book: 1) a
new method to learn online an asymmetric boosted
classifier, pioneering a new direction of online
learning a face detector; 2) a new weak classifier
learning method,
significantly reducing the learning time of a
face detector from weeks to just a few hours; 3) a
new and principled method to learn a
face detector cascade, further improving
the learning time and the detection speed of a face
detector; and 4) a theoretical analysis on the
generalization of an asymmetric boosted classifier
via bounds on the true
asymmetric error of the classifier. The work is
concluded with a discussion of future directions for
face detection.

Autorentext

Minh-Tri Pham is a Research Fellow in computer science at theSchool of Computer Engineering, Nanyang Technological University,Singapore. Tat-Jen Cham is an Associate Professor in computerscience and the Director of the Centre for Multimedia and NetworkTechnology, School of Computer Engineering, Nanyang TechnologicalUniversity, Singapore.

Klappentext

Asymmetric boosting, while acknowledged to beimportant to state-of-the-art face detection, istypically based on the trial-and-error practice,rather than on principled methods. This work solves anumber of issues related to asymmetric boosting andthe use of asymmetric boosting in face detection. Itshows how a proper understanding and use ofasymmetric boosting leads to significant improvementsin thelearning time, the learning capacity, the detectionspeed and the detection accuracy of a face detector.There are four main contributions in this book: 1) anew method to learn online an asymmetric boostedclassifier, pioneering a new direction of onlinelearning a face detector; 2) a new weak classifierlearning method,significantly reducing the learning time of aface detector from weeks to just a few hours; 3) anew and principled method to learn aface detector cascade, further improvingthe learning time and the detection speed of a facedetector; and 4) a theoretical analysis on thegeneralization of an asymmetric boosted classifiervia bounds on the trueasymmetric error of the classifier. The work isconcluded with a discussion of future directions forface detection.