GPU Implementation of Real-Time Biologically Inspired Face Detection using CUDA

In this paper massively parallel real-time face detection based on a visual attention and cortex-like mechanism of cognitive vision system is presented. As a first step, we use saliency map model to select salient face regions and HMAX C1 model to extract features from salient input image and then apply mixture of expert neural network to classify multiview faces from nonface images. The saliency map model is a complex concept for bottomup attention selection that includes many processes to find face regions in a visual science. Parallel real-time implementation on Graphics Processing Unit (GPU) provides a solution for this kind of computationally intensive image processing. By implementing saliency map and HMAX C1 model on a multi-GPU platform using CUDA programming with memory bandwidth, we achieve good performance compared to recent CPU. Running on NVIDIA Geforce 8800 (GTX) graphics card at resolution 640×480 detection rate of 97% is achieved. In addition, we evaluate our results using a height speed camera with other parallel methods on face detection application.

[1]  Christof Koch,et al.  Modeling attention to salient proto-objects , 2006, Neural Networks.

[2]  D. Strayer,et al.  Cell phone-induced failures of visual attention during simulated driving. , 2003, Journal of experimental psychology. Applied.

[3]  Kurt Debattista,et al.  A GPU based saliency map for high-fidelity selective rendering , 2006, AFRIGRAPH '06.

[4]  Reza Ebrahimpour,et al.  Teacher-Directed Learning in View-Independent Face Detection with Mixture of Experts Using Cortex-Like Features , 2010, Aust. J. Intell. Inf. Process. Syst..

[5]  Thomas Serre,et al.  A Theory of Object Recognition: Computations and Circuits in the Feedforward Path of the Ventral Stream in Primate Visual Cortex , 2005 .

[6]  Christof Koch,et al.  A Model of Saliency-Based Visual Attention for Rapid Scene Analysis , 2009 .

[7]  Tingting Xu,et al.  Looking at the surprise: Bottom-up attentional control of an active camera system , 2008, 2008 10th International Conference on Control, Automation, Robotics and Vision.

[8]  Gordon Cheng,et al.  Distributed visual attention on a humanoid robot , 2005, 5th IEEE-RAS International Conference on Humanoid Robots, 2005..

[9]  Thomas Serre,et al.  Object recognition with features inspired by visual cortex , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[10]  David G. Lowe,et al.  Multiclass Object Recognition with Sparse, Localized Features , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[11]  Christof Koch,et al.  Comparison of feature combination strategies for saliency-based visual attention systems , 1999, Electronic Imaging.

[12]  Donald P. Greenberg,et al.  Spatiotemporal sensitivity and visual attention for efficient rendering of dynamic environments , 2005, TOGS.

[13]  Laurent Itti,et al.  Applying computational tools to predict gaze direction in interactive visual environments , 2008, TAP.

[14]  Thomas Serre,et al.  Robust Object Recognition with Cortex-Like Mechanisms , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Tingting Xu,et al.  The autonomous city explorer project: aims and system overview , 2007, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[16]  T. Poggio,et al.  Hierarchical models of object recognition in cortex , 1999, Nature Neuroscience.