A guide on how to take a good selfie that others will like:<p><pre><code> be female
be blonde
be attractive
</code></pre>
Incidentally, Christian Rudder did a really good "study" on the dating site pictures a few years ago:<p><a href="http://blog.okcupid.com/index.php/dont-be-ugly-by-accident/" rel="nofollow">http://blog.okcupid.com/index.php/dont-be-ugly-by-accident/</a>
This is neat. I bet Facebook or OkCupid are sitting on all sorts of click data that could be used to develop tools for helping people make their photos look better. (Even if, personally, I can't wait for a cultural backlash against internet narcissism...)<p>[Edit: Even better, he didn't use click data to train the model, just public likes.]
I would like to see a deep dream selfie ...<p>Feed it an initial picture (noise, clouds, a selfie) and then backwards manipulate the input to maximize the assessed quality of the "selfie".<p>I guess that would look pretty funny.
One thing I always found interesting is Lecun is credited with developing covnets, but Hinton is apparently credited with scaling them and showing the world how great they are in the paper from 2012 - why was Hinton's group (Toronto) able to publish these ground breaking results before Lecun's group (NYU)
>Be female. Women are consistently ranked higher than men. In particular, notice that there is not a single guy in the top 100.<p>This sounds true, but it can't be the real reason—selfies are ranked relative to the other images by the <i>same user</i>. So unless users are taking a lot of #selfies of people of different genders, we can assume the dataset is already controlled for the gender of the person in the image, no? Unless there's some confounding factor at play, such as some demographic segment being more likely to optimize for good selfies occasionally but have boring feeds the rest of the time.<p>would be super interesting, if the data is available, to normalize this by exposure. Of the people that saw an image, how many clicked "like"?
How to take a good selfie: don't be black or dark-skinned, unless you're a celebrity.<p>How do we prevent our AIs from learning racism?<p>EDIT> Informative article, BTW. A good read.
I think it's less about the head getting chopped than about having "the head take up about 1/3 of the image," as Karpathy says. So what the net is learning is composition, or balance in an image, which is really cool. The rule of thirds is actually pretty well know to people in photography:<p><a href="https://en.wikipedia.org/wiki/Rule_of_thirds" rel="nofollow">https://en.wikipedia.org/wiki/Rule_of_thirds</a><p>(Our deep-learning framework <a href="http://deeplearning4j.org" rel="nofollow">http://deeplearning4j.org</a> missed his list, but it's got working convnets, too.)
One caveat with these machine inspired knowledge: they are prone to error, probably more than humans, at least for now.<p>For example, if you train a CNN directly with human faces, its recognition rate comes way below what a human is capable of. Only after you apply tons of handcrafted optimizations, which are mostly black art, will you get close to or surpass a human's capability. Without much domain specific tuning, an AI's insight is far from reliable.
It seems this neural network has a sense of humor if you look at the "Finding the Optimal Crop for a selfie" area.<p>You can see it optimized the last selfie by cropping the face fully out of the picture.. :))
DNN is a key technology of the future. I highly recommend the education program Professor Karpathy mentions at the end of this post. All are excellent and free.
I have seen similar results before: <a href="https://medium.com/the-physics-arxiv-blog/the-algorithm-that-sees-beauty-in-photographic-portraits-435ab8064646" rel="nofollow">https://medium.com/the-physics-arxiv-blog/the-algorithm-that...</a>