Hi Edouardo,

Thanks for the comments

Which one do you call the "preliminary" sketch?

All the images in the galleries on the site are produced of Aikon-1 which
was a my MSc final project (2006). They would be better suited to produce
etchings. It is a fairly "stupid" program. The more recent recent version
of aikon-i that produces things like the image in the new scientist is
from 2008, I still use it because it always gives a result and some are
quite interesting when rendered at a certain scale. And people like them,
so they get published (Blueprint, New Scientist...). Aikon-i wasn't an
investigation into sketching. The idea was to implement something inspired
by a strategy I used when I was still sketching and see what happens. I
never try to get a specific result: I implement some processes and see
what happens...The intention even with aikon-ii is not to get a result
looking like a drawing made by a human.

Aikon-ii is a totally different system.
Aikon-ii is just starting to do things. There is a video on the website of
what it is doing so far, it only started to draw last February. It will
draw using lines and look at what it is drawing. I am working on it now...

I'll try to have a look at "The Natural Way To Draw"

Best

Patrick




On Wed, April 7, 2010 11:09 am, Eduardo Corte Real wrote:
> Hi Patrick,
> Thanks for the information.
> I found far more interesting the "preliminary" sketch than the final
> ones. That kind of lining in the final ones is common to sculptors and was
> also suggested by Kimon Nicolaïdes in his "Natural Way to Draw". It was
> mastered by Henry Moore and Giacometti among others as a way to achieve a
> three-dimensional  perception but Aikon´s results flatten the image. I'm
> sure you know the article: Pearson, D., Hanna, E. and Martinez, K. (1990)
> Computer-generated cartoons. In: /Images and Understanding/, pp.
> 46-60, Cambridge University Press. I don't know far we have gone from
> there, but identifying darker areas and make a correspondent rendering
> (even if apparently more natural like aikon's lining don't seem to be
> the way. Also I'm curious about one thing: does Aikon "sees" the drawing
> while making it or only sees the model? Best,
> Eduardo
> PS: we could discuss the "Natural Way to Draw" as a good starting point
> because being a postulate.
>
> On 06-04-2010 19:11, Patrick Tresset wrote:
>
>> Dear Eduardo,
>>
>>
>> The New Scientist article doesn't represent my views on our research
>> project. The text on our website is far more accurate.
>> http://www.aikon-gold.com
>>
>>
>> Best
>>
>>
>> Patrick
>>
>>
>>
>> Patrick Tresset
>> Co-Lead, Aikon Project
>>
>>
>> On Tue, April 6, 2010 5:12 pm, Eduardo Corte Real wrote:
>>
>>
>>> Dear List,
>>>
>>>
>>>
>>> in another List "PhD Design List" there is an interesting thread
>>> resulting from a post by Terence Love about robot drawing:
>>> http://www.newscientist.com/blogs/culturelab/2010/04/life-drawing-rob
>>> ot-c ould-teach-us-about-art.php.
>>>
>>> Any thoughts?
>>> Eduardo Corte-Real
>>> Doctor Arch, IADE Lisbon
>>>
>>>
>>>
>>>
>>
>>
>


--