Best Selfies

Does A Deep Neural Network Like Your #Selfie?

Best Selfies
Andrej Karpathy trained a Convolutional Neural Network with a dataset of 2 million photographs to determine what makes the perfect selfie. The image above contains the top 100 best selfies (here are the 1,000 best selfies) Andrej concludes that the best selfies have these qualities:

  1. Be female. Women are consistently ranked higher than men. In particular, notice that there is not a single guy in the top 100.
  2. Face should occupy about 1/3 of the image. Notice that the position and pose of the face is quite consistent among the top images. The face always occupies about 1/3 of the image, is slightly tilted, and is positioned in the center and at the top. Which also brings me to:
  3. Cut off your forehead. What’s up with that? It looks like a popular strategy, at least for women.
  4. Show your long hair. Notice the frequent prominence of long strands of hair running down the shoulders.
  5. Oversaturate the face. Notice the frequent occurrence of over-saturated lighting, which often makes the face look much more uniform and faded out. Related to that,
  6. Put a filter on it. Black and White photos seem to do quite well, and most of the top images seem to contain some kind of a filter that fades out the image and decreases the contrast.
  7. Add a border. You will notice a frequent appearance of horizontal/vertical white borders.

Andrej also created a TwitterBot that will judge your selfie. Simply attach your selfie (or a include a link) to a tweet that mentions @deepselfie anywhere in it. The bot will analyze your selfie and give you its opinion (e.g. score 90% means that the Selfie Bot is 90% sure yours would be in top half of selfies. Selfie Bot was not impressed with my selfie.


More here.

Noteworthy Tumblr #5: Table For One

Table For One is simple. It has a white background, photos of people eating alone, a title, and that’s about it. However, it elicits some strong feelings from people.

The tumblr didn’t evoke any negative emotions from me, maybe because I eat alone so often. I enjoy eating by myself. You can’t forget that just because you are at a table for one, it doesn’t mean you are alone.


Photo credit: Jerry Hsu

Andrew WK: Harm

Despite all the rumors Andrew WK is human after all. Or at least his persona is had the all-too-familiar angst/lust/confusion of a typical 17 year old. He recently published an article in the Gaurdian about obsessive feelings he had towards his girlfriend and how he ended up expressing those feelings through song. A song that ended up leading to a juvenile restraining order.

It turned out that the assistant to the head of school got a copy of my song on cassette and gave it to the girl I had a crush on. This was probably the worst thing that had ever happened in my life. She heard the song and was completely freaked out. Within three days, every kid in school had a copy. She told her friends, teachers and parents: “This guy at school is stalking me and threatening my life.” She played them the song and they called the police.

The song is actually pretty good for a crazed 17 year-old. Have a listen (The lyrics can be found below th fold) and then go read Andrew’s full account of the embarrassing event:

Andrew WK – Harm

Also, yes I’m very aware that this story is likely not true and simply part of the Andrew WK “project” (marketing machine) and is simply just a story. For that, I may actually appreciate it even more.
Read More

One Year’s Addictions

I love this project put together by Rollertrain. “One Year’s Addictions” is a collection of one entire years worth of “empty cigarette boxes, tubes of chapstick, beer bottle tops, empty bottles of lubricant, empty prescription pill bottles and broken ipod headphones.” I shutter to think of what my collection of addictions would look like – it would be very different, no doubt, but no less sinister.

One years addiction
Photo courtesy of Libby Lynn.