The future of the computation is the Graphical Processing Unit, i. e. the GPU. The graphics cards have been shown in the pasture of image processing and accelerated interpretation of 3D scenes, and computational capability that these GPUs acquire, they are rising into immense parallel computing units. It is quite simple to program any graphics processor to execute universal parallel tasks. But after understanding the various architectural features of the graphics processor, it can be used to execute other demanding tasks as well. In this paper, the use CUDA (Compute Unified Device Architecture) can fully utilize most tremendous power of these GPUs. CUDA is most popular parallel computing architecture of NVIDIA. It enables dramatic increases in computing performance, by harnessing the power of the GPU. In this, the sequential software of the famous image processing problem, the pedestrian detection[1] and parallelize it to run on GPUs and increase the performance of that software. We have presented a new pedestrian detector that improves in speed and quality of the pedestrian detection. By efficiently handling different scales and transferring computation from the test time to training time, detection speed is improved.

1. N. Dalal and B. Triggs. Histograms of Oriented Gradients for Human Detection. In Proc. of the IEEE Conference on Compute Vision and Pattern Recognition, pages 886–893, 2005.
2. S. H. U. Chang, D. Xiaoqing, and F. Chi, Histogram of the Oriented Gradient for Face Recognition , Tsinghua Science and Technology, vol. 16, 2011, no. 2, pp. 216-224.
3. R. Benensonetal. Pedestrian detection at 100 frames per second. In CVPR, 2012. P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan. Object detection with discriminately trained part-based models. Pattern Analysis and Ma-chine Learning (PAMI), Sept. 2010.
4. NVIDIA. NVIDIA's next generation CUDA computer architecture:Kepler GK110. http://www. nvidia. com/, 2012.
5. A. Danalis, G . Marin, C. McCurdy, J. K. Spafford, V. Tipparaju, and J. S. Meredith, P. C. Roth, S. Vetter. T he Scalable Heterogeneous Computing ( S H O C) benchmark suitings of the 3rd In Proceed Workshop o n General-Purpose Computation on Graphics Processing Units, GPGPU ' 1 0, pages 63-74, New York, NY, USA, Mar. 2 0 1 0 . ACM
6. K. Mizunoetal. Architectural Study of HOG Feature Extraction Processor for Real-Time Object Detection. In SiPS, 2012.
7. P. Viola, M. J. Jones, and D. Snow. Detecting pedestriansusing patterns of motion and appearance. The 9th ICCV, Nice, France, volume 1, pages 734–741, 2003.
8. Z. Song, M. Wang, X. Hua, and S. Yan. Predicting occupation via human clothing and contexts. In ICCV, 2011.
9. P. Dollár, C. Wojek, B. Schiele, and P. Perona. Pedestrian detection: An evaluation ofthe state of the art. PAMI, 34(4):743–761, 2012.

Pedestrians Detection Compute Unified Device Architecture(cuda) Object Detection Graphics Processing Units (gpus) Parallel Architectures svm Classifier.