Title :
Effective false positive reduction in multilevel face detection system using convolutional neural networks
Author :
Timoshenko, Denis ; Grishkin, Valery ; Smirnov, Evgueni
Author_Institution :
St.-Petersburg State Univ., St. Petersburg, Russia
fDate :
June 30 2014-July 4 2014
Abstract :
In this paper, the robust method for reduction of `false positive´ in<;br/>multilevel face detection system is proposed. The method is based on<;br/>convolutional neural network and trained on results from cascades of<;br/>boosted classifiers. Moreover, we evaluate the effect of modern<;br/>approaches like Dropout and DropConnect on network learning.
Keywords :
face recognition; feature extraction; learning (artificial intelligence); neural nets; convolutional neural networks; false positive reduction; multilevel face detection system; network learning; Decision trees; Detectors; Educational institutions; Face; Face detection; Neural networks; Robustness;
Conference_Titel :
Computer Technologies in Physical and Engineering Applications (ICCTPEA), 2014 International Conference on
Conference_Location :
St. Petersburg
Print_ISBN :
978-1-4799-5315-8
DOI :
10.1109/ICCTPEA.2014.6893351
