Interesting - with suitably chosen distortions and colours these could be
turned into faces I guess - the main point is to produce things such that
someone choosing one of them will be able to pick the same one again
a week or so later - and they will be jumbled up - so remembering
"it's the first one" will be no use.

I just assume that people are probably going to be best at remembering
a persons face, rather than abstract stuff - hence the idea to use faces,
but anything else that's rememberable would also suit I expect.

Chris.

"JC (Exether)" <no### [at] spamfr> wrote in message
news:3f2e8431$1@news.povray.org...
> I just read the offer and it seems that it is expected to output
> concrete images (objects, cats, dogs, etc.), not abstract. But you can
> argue that having abstract images makes it even harder to describe to
> someone else.
> Beyond that I think your images are too similar, It would be very
> difficult to pick the right one without a lot of training. Maybe you
> should vary them a little bit more, changing the colors for example.
>
> JC
>
> Micha Riser wrote:
> > see p.unix "Paid Project offer (image gneration for security purposes)"
> >
> >
> > ------------------------------------------------------------------------
> >
>

Post a reply to this message