Enhancing the Performance of GPU for Face Detection
H. M. Faheem, Said Ghoniemy, T. A. Yara M. Abdelaal · International Journal of Computer Applications · 2014
Computer Vision algorithms are considered computationally intensive problems.Face detection is one of the most complex objects to detect due to its variations.The objective is to enhance the face detection time (compared with other approaches) to reach a real-time application that will be later on used in augmented reality applications such as telepresence.The experiments with NVIDIA GTX 560 show that detecting the faces in an image of size [640x480] can process up to 33 frames per second, also this paper shows how the researcher's approach can be generalized to support larger image sizes.This in turn reflects back the achieved speed that exceeds FPGA.