A Fast and Efficient SIFT Detector Using the Mobile GPU
Department of Electrical and Computer Engineering, Rice University, Houston, Texas
IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2013
@article{rister2013fast,
title={A FAST AND EFFICIENT SIFT DETECTOR USING THE MOBILE GPU},
author={Rister, Blaine and Wang, Guohui and Wu, Michael and Cavallaro, Joseph R},
year={2013}
}
Emerging mobile applications, such as augmented reality, demand robust feature detection at high frame rates. We present an implementation of the popular Scale-Invariant Feature Transform (SIFT) feature detection algorithm that incorporates the powerful graphics processing unit (GPU) in mobile devices. Where the usual GPU methods are inefficient on mobile hardware, we propose a heterogeneous dataflow scheme. By methodically partitioning the computation, compressing the data for memory transfers, and taking into account the unique challenges that arise out of the mobile GPU, we are able to achieve a speedup of 4-8x over an optimized CPU version, and a 6.4x speedup over a published GPU implementation. Additionally, we reduce energy consumption by 87 percent per image. We achieve near-realtime detection without compromising the original algorithm.
February 28, 2013 by hgpu
