How an A.I. ‘Cat-and-Mouse Game’ Generates Believable Fake Photos

The New York Times is running a very fascinating article on the progress of the artificial intelligence and machine learning in both identifying and generating fake photos – How an A.I. ‘Cat-and-Mouse Game’ Generates Believable Fake Photos.   The above image shows the progress of the AI working against itself and learning from its own results – one part is trying to identify if the photo is fake or not, and the other part is trying to generate a fake photo which will pass the test.  When the test fails, the system learns, improves, and tries again.  Look at the last row of photos, which are super realistic and took the system between 10 to 18 days to learn how to generate.

But that’s not all.  It gets better, and I quote:

A second team of Nvidia researchers recently built a system that can automatically alter a street photo taken on a summer’s day so that it looks like a snowy winter scene. Researchers at the University of California, Berkeley, have designed another that learns to convert horses into zebras and Monets into Van Goghs. DeepMind, a London-based A.I. lab owned by Google, is exploring technology that can generate its own videos. And Adobe is fashioning similar machine learning techniques with an eye toward pushing them into products like Photoshop, its popular image design tool.

Here are a few more photos that were generated:

This is remarkable.  But if you keep reading the article, you’ll quickly discover that there is even more to it.  What’s next in line after pictures?  You are correct: videos.  You better sit down before you watch this video, showing Obama’s lip sync:

So, can’t trust the TV.  Can’t trust the Internet.  Who do you trust?

Defensive Programming : Object Calisthenics

I came across this nice and somewhat strongly opinionated video on Defensive Programming:

Marco Pivetta makes quite a few good points with I agree (and a few with which I disagree).  One thing that he mentioned though I haven’t heard about – Object Calisthenics.  Which turns out to be yet another set of rules and best practices for the object-oriented design and programming.  Here are the rules to get you started:

  1. Only One Level Of Indentation Per Method
  2. Don’t Use The ELSE Keyword
  3. Wrap All Primitives And Strings
  4. First Class Collections
  5. One Dot Per Line
  6. Don’t Abbreviate
  7. Keep All Entities Small
  8. No Classes With More Than Two Instance Variables
  9. No Getters/Setters/Properties

Read the whole article for explanations and examples.

Fungal Intelligence

With all the recent hype around artificial intelligence, this thing that I came across today is a breath of fresh air.  The subject is: fungal intelligence.  “WTF?”, I hear you ask.  Have a look at this tiny video:

Crazy, right?  Well, crazy enough to be found in this article at Nature.com.  And, how the heck did I end up there?  That’s the first link in the Google search results for “fungal intelligence”, which I had to look up after watching this video:

Fascinating stuff!

Charlottesville: Race and Terror

Charlottesville was a lot in the news recently.  I didn’t pay much attention, but now I see why.  This is crazy.  It almost feels unreal, like a really long trailer or a promotion video to a new movie.  But it’s not.  It’s real life and it’s happening now.

It’s far from funny, but standup comedians are often some of the smartest people, with excellent observation skills and the unbeatable use of words.  So here’s Jim Jefferies take on this, with which I agree wholeheartedly.

Replacing default MATE window manager with i3wm

For the last few month I’ve been torn between i3 tiling window manager and MATE desktop environment.  I love them both, even though they are very different from each other.

When I’m alone, and have all the time in the world to write code, i3wm is my best friend.  When I’m in the office, or need to move between tasks, monitors, and locations, MATE is the best.  But I want both.  I don’t want to choose.

Today I came across this YouTube video with a screencast of how to setup i3 window manager instead of the default MATE’s one.  Yes, i3 running inside MATE!  This sounds like magic!

Things I have to do this week make tweaking a working desktop a really bad idea, so I probably won’t try this until the weekend, but it does look exciting!

X27 – ultra low light night vision color camera

As someone who works with technology for years now, I might sometimes appear as a “know all, seen all” kind of guy.  But that’s far from the truth.  Sure, there is plenty of technology I know or even slightly aware of.  But there is still a whole lot that blows my mind when I come across it.

Via this blog post, I came across an X27 – ultra low light night vision color camera.  And my mind is blown.  This technology provides high definition, color video images in ultra low light conditions (no moon, overcast).  The level of detail this camera picks up is simply unbelievable.  Check out this video comparing X27 to some other night vision technologies.

The X27 camera takes videos in darkness that looks like they were shot in the daytime. And they’re in color…none of this black and white, thermal, or infrared stuff. The camera was developed for military use, has an effective ISO rating of 5,000,000, and has a comically long name: “X27 Reconnaissance Day/Night high Fidelity true real time low light/low lux color night vision Imaging Security / Multi Purpose camera system”. Pricing information is not available, but I bet you’re paying for every single one of those words. (via digg)

Sure, military is probably the first and only sector that can afford this.  But as with anything, once the technology is available, it will make it’s way down to the rest of us, opening new doors in law enforcement, security, photography, movies and TV, entertainment, health and safety, environment and animal welfare, and other areas.

Super cool stuff!