GPU Vision: Accelerating Computer Vision algorithms with Graphics Processing Units
(2011)
@article{lengyel2011gpu,
title={GPU Vision: Accelerating Computer Vision algorithms with Graphics Processing Units},
author={Lengyel, T.K. and Gedarovich, J. and Cusano, A. and Peters, T.J.},
year={2011}
}
We present an introduction to the field of GPU accelerated computer vision by examining several projects that provide the framework for researchers and developers to tap into the computational power of Graphics Processing Units (GPU). Our goal is to identify the tools and areas where GPU acceleration can provide the highest performance increases in computer vision applications by creating performance benchmarks to compare and contrast the GPU and CPU versions in realistic applications. We focus our attention on open source projects, such as GpuCV and OpenCV’s upcoming GPU module that are compatible with NVIDIA’s CUDA programming language. Additionally, we examine a research project that ported the Viola-Jones face-detection algorithm to CUDA.
March 20, 2011 by hgpu
