Minh Tri Pham (4 results)

Taschenbuch. Condition: Neu. Face Detection with Asymmetric Boosting | Principled Methods to Rapid Learning and Classification | Minh-Tri Pham | Taschenbuch | Englisch | VDM Verlag Dr. Müller | EAN 9783639178326 | Verantwortliche Person für die EU: preigu GmbH & Co. KG, Lengericher Landstr. 19, 49078 Osnabrück, mail[at]preigu[dot]de | Anbieter: preigu. …

paperback. Condition: Like New. LIKE NEW. SHIPS FROM MULTIPLE LOCATIONS. book.

Kartoniert / Broschiert. Condition: New. Dieser Artikel ist ein Print on Demand Artikel und wird nach Ihrer Bestellung fuer Sie gedruckt. Autor/Autorin: Pham Minh-TriMinh-Tri Pham is a Research Fellow in computer science at the School of Computer Engineering, Nanyang Technological University, Singapore. Tat-Jen Cham is an Associate Professor in computer science and the Director of th. …

Taschenbuch. Condition: Neu. nach der Bestellung gedruckt Neuware - Printed after ordering - 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.…