MIT and Harvard researchers demonstrate a better way for computers to see

Journal article (open access): <a href="http://www.ploscompbiol.org/article/info%3Adoi%2F10.1371%2Fjournal.pcbi.1000579" rel="nofollow">http://www.ploscompbiol.org/article/info%3Adoi%2F10.1371%2Fj...</a><p>It's a really neat application of General Purpose GPU programming:<p><a href="http://gpgpu.org/" rel="nofollow">http://gpgpu.org/</a><p><a href="http://en.wikipedia.org/wiki/GPGPU" rel="nofollow">http://en.wikipedia.org/wiki/GPGPU</a>

Granted I just skimmed the article, but do they discuss the algorithm at all?? It seemed like they're just saying buzzwords, "parallel","gpu","vision is hard", etc.<p>Does anyone know if a new CV algorithm is involved in this?