A Comparison of FPGA and GPU for Real-Time Phase-Based Optical Flow, Stereo, and Local Image Features
Issue No. 07 - July (2012 vol. 61)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TC.2011.120
Karl Pauwels , K.U.Leuven, Leuven
Matteo Tomasi , University of Granada, Granada
Javier Díaz , University of Granada, Granada
Eduardo Ros , University of Granada, Granada
Marc M. Van Hulle , K.U.Leuven, Leuven
Low-level computer vision algorithms have extreme computational requirements. In this work, we compare two real-time architectures developed using FPGA and GPU devices for the computation of phase-based optical flow, stereo, and local image features (energy, orientation, and phase). The presented approach requires a massive degree of parallelism to achieve real-time performance and allows us to compare FPGA and GPU design strategies and trade-offs in a much more complex scenario than previous contributions. Based on this analysis, we provide suggestions to real-time system designers for selecting the most suitable technology, and for optimizing system development on this platform, for a number of diverse applications.
Reconfigurable hardware, graphics processors, real-time systems, computer vision, motion, stereo.
K. Pauwels, M. Tomasi, E. Ros, M. M. Van Hulle and J. Díaz, "A Comparison of FPGA and GPU for Real-Time Phase-Based Optical Flow, Stereo, and Local Image Features," in IEEE Transactions on Computers, vol. 61, no. , pp. 999-1012, 2011.