Implementation of face selective attention model on an embedded system

Authors:
Bumhwi Kim;Hyung-Min Son;Yun-Jung Lee;Minho Lee
Affiliations:
School of Electrical Engineering and Computer Science, Kyungpook National University, Daegu, South Korea;School of Electrical Engineering and Computer Science, Kyungpook National University, Daegu, South Korea;School of Electrical Engineering and Computer Science, Kyungpook National University, Daegu, South Korea;School of Electrical Engineering and Computer Science, Kyungpook National University, Daegu, South Korea
Venue:
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part V
Year:
2012

Citing 9
Cited 0

A Model of Saliency-Based Visual Attention for Rapid Scene Analysis

IEEE Transactions on Pattern Analysis and Machine Intelligence
Saliency, Scale and Image Description

International Journal of Computer Vision
Stochastic Guided Search Model for Search Asymmetries in Visual Search Tasks

BMCV '02 Proceedings of the Second International Workshop on Biologically Motivated Computer Vision
Visual Attention Using Game Theory

BMCV '02 Proceedings of the Second International Workshop on Biologically Motivated Computer Vision
Robust Real-Time Face Detection

International Journal of Computer Vision
2008 Special Issue: Stereo saliency map considering affective factors and selective motion analysis in a dynamic environment

Neural Networks
Improving AdaBoost Based Face Detection Using Face-Color Preferable Selective Attention

IDEAL '08 Proceedings of the 9th International Conference on Intelligent Data Engineering and Automated Learning
Face detection with the modified census transform

FGR' 04 Proceedings of the Sixth IEEE international conference on Automatic face and gesture recognition
Affective saliency map considering psychological distance

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a new embedded system which can selectively detect human faces with fast speed. The embedded system is developed by using OMAP 3530 application processor which has DSP and ARM core. Since the embedded system has the limited performance of CPU and memory, we propose a hybrid system combined the YCbCr based bottom-up selective attention with the conventional Adaboost algorithm. The proposed method using the bottom-up selective attention model can reduce not only the false positive error ratio of the Adaboost based face detection algorithm but also the time complexity by finding the candidate regions of the foreground and reducing the regions of interest (ROI) in the image. The experimental results show that the implemented embedded system can successfully work for localizing human faces in real time.