Compact convolutional neural network cascadefor face detection

Bibliographic Details
Parent link:CEUR Workshop Proceedings: Online Proceedings for Scientific Conferences and Workshops
Vol. 1576 : Parallel Computing Technologies 2016, PCT 2016.— 2016.— [P. 375-387]
Main Author: Kalinovsky I. A. Iljya Andreevich
Corporate Author: Национальный исследовательский Томский политехнический университет (ТПУ) Институт кибернетики (ИК) Кафедра вычислительной техники (ВТ)
Other Authors: Spitsyn V. G. Vladimir Grigorievich
Summary:Title screen
This paper presents a new solution to the frontal face detection problem based on a compact convolutional neural networks cascade. Test results on an FDDB dataset show that it is able to compete with state-of-the-art algorithms. This proposed detector is implemented using three technologies: SSE/AVX/AVX2 instruction sets for Intel CPUs, Nvidia CUDA, and OpenCL. The detection speed of our approach exceeds considerably all the existing CPUbased and GPU-based algorithms. Thanks to its high computational efficiency, our detector can process 4K Ultra HD video stream in real time (up to 27 fps) on mobile platforms while searching objects with a dimension of 60×60 pixels or higher. At the same time, its processing speed is almost independent of the background and the number of objects in a scene. This is achieved by asynchronous computation of stages in the cascade.
Language:English
Published: 2016
Subjects:
Online Access:http://earchive.tpu.ru/handle/11683/36142
http://ceur-ws.org/Vol-1576/150.pdf
Format: Electronic Book Chapter
KOHA link:https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=650721

MARC

LEADER 00000naa0a2200000 4500
001 650721
005 20250903132357.0
035 |a (RuTPU)RU\TPU\network\15970 
090 |a 650721 
100 |a 20161018d2016 k||y0engy50 ba 
101 0 |a eng 
102 |a DE 
135 |a drcn ---uucaa 
181 0 |a i  
182 0 |a b 
200 1 |a Compact convolutional neural network cascadefor face detection  |f I. A. Kalinovsky, V. G. Spitsyn 
203 |a Text  |c electronic 
300 |a Title screen 
330 |a This paper presents a new solution to the frontal face detection problem based on a compact convolutional neural networks cascade. Test results on an FDDB dataset show that it is able to compete with state-of-the-art algorithms. This proposed detector is implemented using three technologies: SSE/AVX/AVX2 instruction sets for Intel CPUs, Nvidia CUDA, and OpenCL. The detection speed of our approach exceeds considerably all the existing CPUbased and GPU-based algorithms. Thanks to its high computational efficiency, our detector can process 4K Ultra HD video stream in real time (up to 27 fps) on mobile platforms while searching objects with a dimension of 60×60 pixels or higher. At the same time, its processing speed is almost independent of the background and the number of objects in a scene. This is achieved by asynchronous computation of stages in the cascade. 
461 |t CEUR Workshop Proceedings  |o Online Proceedings for Scientific Conferences and Workshops 
463 |t Vol. 1576 : Parallel Computing Technologies 2016, PCT 2016  |o Proceedings of the 10th Annual International Scientific Conference on Parallel Computing Technologies, Arkhangelsk, Russia, March 29-31, 2016  |v [P. 375-387]  |d 2016 
610 1 |a электронный ресурс 
610 1 |a труды учёных ТПУ 
610 1 |a распознавание лиц 
610 1 |a нейронные сети 
610 1 |a классификаторы 
700 1 |a Kalinovsky  |b I. A.  |c specialist in the field of Informatics and computer engineering  |c assistant at Tomsk Polytechnic University  |f 1990-  |g Iljya Andreevich  |3 (RuTPU)RU\TPU\pers\37438 
701 1 |a Spitsyn  |b V. G.  |c specialist in the field of informatics and computer technology  |c Professor of Tomsk Polytechnic University, Doctor of technical sciences  |f 1948-  |g Vladimir Grigorievich  |3 (RuTPU)RU\TPU\pers\33492  |9 17160 
712 0 2 |a Национальный исследовательский Томский политехнический университет (ТПУ)  |b Институт кибернетики (ИК)  |b Кафедра вычислительной техники (ВТ)  |3 (RuTPU)RU\TPU\col\18699 
801 2 |a RU  |b 63413507  |c 20170124  |g RCR 
856 4 |u http://earchive.tpu.ru/handle/11683/36142 
856 4 |u http://ceur-ws.org/Vol-1576/150.pdf 
942 |c CF