Facebook Uses Artificial Intelligence to Generate Realistic Virtual Imagery

Facebook Uses Artificial Intelligence to Generate Realistic Virtual Imagery

Reading Time: 2 minutes

Facebook Uses Artificial Intelligence to Generate Realistic Virtual Imagery 

This has been a big few weeks for machine learning and imagery.(See links below). Now Facebook just published new research outlining a novel approach for generating realistic, artificial images of scenery and things like dogs, planes, deer, ships, trucks, horses, and, of course, cats.

I’ve tried parsing through the research, but the details are beyond me. What is interesting, however is their use of an approach called “Generative Adversarial Networks” (GAN). Essentially, what they’ve done is create a kind of feedback loop between two networks, where the first, the “generative network” generates an image from noise. Then the other, “discriminative network,” takes that resulting image, and essentially compares it to training data that is based on real images (note: this is a slight simplification). The result is that with each iteration, the generative network is ‘tricked’ into generating increasingly realistic looking imagery. 

The researchers then testing the resulting images with a group of volunteers and found that 40% of the images were realistic enough to fool a human into thinking they are real images.

What is Facebook likely to do with the results of this research? That’s unclear, but with their Oculus Virtual Reality acquisition, it seems reasonable to assume that they are going to need cost-effective methods for generating a massive scale of virtual scenery and objects. Could this research represent early forays into that work?  

Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks

http://arxiv.org/pdf/1506.05751v1.pdf

* More:

Looking Inside the Image Recognition of Artificial Intelligence:

https://plus.google.com/u/0/+GideonRosenblatt/posts/gQw9kP8CKzY

Is This the First Computational Imagination?

https://plus.google.com/u/0/+GideonRosenblatt/posts/8V82FxXKxXD

#artificialintelligence   #machinelearning   #virtualreality   #facebook  

No comments

  1. New seeing, new imagining. Familiar, of course, we’ve all been there.

  2. Wow! What a long strange trip these last two “big weeks” have been, Gideon Rosenblatt. I just finished adding into Evernote a research article from Google: “Inceptionism: Going Deeper into Neural Networks” (http://googleresearch.blogspot.com/2015/06/inceptionism-going-deeper-into-neural.html). Far less academic in tone and style than Facebook’s paper, instead of using neural networks to generate more realistic images, Google’s network created far more psychedelic impressions out of random noise.

    Realistic or psychedelic, what would these patterns of bits signify without a human imagination to interpret and enjoy them?

  3. Yes, Leland LeCuyer, that Google research is very interesting also, and there’s research along these lines out of Japan too (published by MIT). Interesting that this is all happening right now.

    And yes, none of this would be happening w/out the human initiative and capacity to appreciate them. Will that always be the case? Probably not. 

  4. “Will that always be the case?” implies a long, longggg time. But if it were to happen at all, I’m less confident than you that it will be realized within a normal human time scale. For the foreseeable future, say the next thousand years, I believe that living animals will the only entities capable of appreciating art or nature. But I’ve been surprised before.

  5. Only Time will tell

  6. เราอ่านไม่ออกเป็นภาษาไทยอย่างเดืยว

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up here for the latest articles. You can opt out at any time.


Subscribe by email:



Or subscribe by RSS:

Subscribe-by-RSS---Orange-Background
%d bloggers like this: