Best Beloved recently pointed out to me that it had been approximately seventeen years since I had recommended anything, which had of course triggered the inevitable thing where I am reluctant to write up a recommendations post because it's been so long since I've done one. You all know how this goes. She suggested I ease myself into it with vids.
(
Read more... )
Comments 34
Elizabeth: You didn't need to. I have eyes.
HEEEE! Your dialogue break-down of Let's Misbehave is to die for, just so you know. Also, I'm probably the only person alive who hasn't seen The Tenth Doctor Musical yet, so I'm going to assume the rec is directed at me personally. ;-)
Reply
Reply
Reply
But, yeah, some of it, BB is there patiently pointing out things and I'm just saying, "Wow. I - I didn't know that was there!" and "Seriously? There's come in this picture?"
(And some of it I'm like OH GOD UNCANNY VALLEY LOOK AWAY. Some photoshopped stuff just looks so scary-wrong-bad to me.)
Reply
Reply
Reply
This should be your new journal name, just sayin'.
Is it all line drawings? Or fanart specifically? Not I want to do SCIENCE on you!
Reply
You might be right about that.
Is it all line drawings? Or fanart specifically? Not I want to do SCIENCE on you!It's - it's hard to explain. It's a style of drawing, and I struggle with it as much in non-fanart as in fanart. It's worst in very detailed, very busy, very line-filled drawings, and worst of all if it's all lines in one color. (So, like, charcoal, sepia, pencil, etc.) I'm just lost; the artist is apparently relying on her audience having an ability I don't have, something that helps people make sense of that kind of jumble ( ... )
Reply
It sounds as though you have a low-quality Face Recognition Module, and it breaks down (almost) completely under unnatural conditions (e.g. line drawings).
What's scientifically interesting is that you *do* see the tiger. (what picture is this? Is it online?)
My hypothesis is that the normal human Face Recognition Module is so advanced that it doesn't need many pixels of input to get a result. You may be evaluating faces as though they are normal patterns.
So if the picture has 100K of data that are part of the tiger, and 20K that are faces, you find the tiger 5 times easier to discern than the faces. The standard human FRM is perhaps 10 times better than that for normal patterns, so to neurotypical person looking at the picture sees the faces twice as well (parses the pattern twice as easily) as the tiger.
... now I want to make your blood relatives look at pictures, too. I get more Agatha Heterodyne-y by the day.
Reply
(Hi TFV! I promise I'm not trying to be creepy; I find your brain interesting both for the witty writing AND the unusual neurological function! But, you know, if you care to tell me whether you find it unusually difficult to guess ages, or whether any of these photos seem strange, you'd make my day!)
Reply
Which is probably another 25, 30% right there.
Reply
Reply
Leave a comment