On Fri, 26 Aug 2005, Richard Clayton wrote:
> Tony Finch <dot@???> writes
>
> >There has also been recent research that shows computers will soon be at
> >least as good as people at solving the current generation of CAPTCHAs.
>
> they're already better :)
> that's assuming that the "segmentation problem" of dividing the graphic
> into constituent characters is not too complex
Which is rather important given that real-world CAPTCHAs have multiple
characters, so I think one should be hesitant to overstate the results of
that paper.
Tony.
--
<fanf@???> <dot@???>
http://dotat.at/ ${sg{\N${sg{\
N\}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}\
\N}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}
