Pareidoloop: Faces generated at random
iobound.com
iobound.com
On the other hand, I vaguely recall a talk by VS Ramachandran where he claimed that in essence that's what Picasso and many others did also, i.e., you might see a painting that objectively doesn't strongly resemble a face, but it hits the right spots in your brain's face detector, so your brain says to itself "now that's what I call a face". A bit like how colorful candy triggers "ripe fruit - eat it!".
They consist of adding and subtracting different rectangles within the image and comparing the sum to a pre-defined threshold (this is called a "weak classifier" in this context). This is done for ~30-50 different weak classifiers and if all of them pass, then a face is declared there.
Therefore, it should be relatively easy to find a set of rectangles that would satisfy these conditions.
My guess, from having worked with face detectors for quite a while, is that it wouldn't be nearly as cool as what painters do -- the false detections would be more mundanely like faces, or so completely off-the-wall if they happen to be a pathological case.
Which could be kind of neat actually... you could train it to look for Jesus for instance.
http://www.artyfactory.com/art_appreciation/animals_in_art/p...
One interesting observation - I turned up the output resolution without changing the target fitness and started noticing that the (larger) results didn't look nearly as much like faces to me. At first I thought there must be a technical reason for this, until I realized that it was all psychological - the smaller images hold less information, and therefore my brain has to try harder to fill in the gaps, tricking me into believing that the low-res faces have more detail than they actually do. To prove this to myself, I scaled down the larger images and - sure enough - the (imagined) faces became much more believable.
Now do it with the automatic porn-filters rather than face recognition ;)
Also, this seems to be a very clear simulation of how we think evolution and natural selection work together. Fun.
With the face example, black-box optimization might be the only practical choice, though, since the face-detection component is probably not easy to express in a nice mathematical form.
I agree it's interesting, if not very practical for anything.
See a sample here : http://i.imgur.com/xmM6U.png
and repo here : https://github.com/vhf/pareidoloop/blob/master/README.md
I would think the faces would tend to look similar, but that's not the case for me so far.
It would be fun to set detection to certain ages and gender; also if a site collected renderings.
The degree of flexibility found in, for example, Second Life would be awesome.
Any leads?
Of course, you need their database to make the algorithm work, but if you emailed the authors they might be accomodating..
Interestingly, and perhaps as clue to your first suggestion, experimentation implies crows can recognize and remember human faces over long durations of time. http://www.sciencedirect.com/science/article/pii/S0003347209...
An interesting study about this[1] shows that children exposed to other races (adopted and moved to another region) lose their initial training.
[1] http://www.pallier.org/papers/SangrigoliPallier_finaldraft.p...