Accelerated Computer Vision in MATLAB
March 7, 2010 by John Melonakos · Leave a Comment
Filed under: Algorithms, Code Optimization, Image processing
Filed under: Algorithms, Code Optimization, Image processing
A recent post on the AccelerEyes’ blog discusses computer vision algorithm acceleration in MATLAB. The key problem in using MATLAB for computer vision is that:
“Matlab and the M language is great for linear algebra where blocks of matrices are the typical access pattern, but not for Computer Vision where algorithms typically operate on patches of imagery.”
In order to get around this problem, a windows function is introduced:
“The command windows signals to Jacket that we’re doing a patched access pattern that can then be optimized on the GPU.”
Anyway, this may be interesting for those involved in computer vision and interested in GPUs!
Popularity: 1% [?]
Translator
Follow Us
Video Posts
