Image Completion with Deep Learning in TensorFlow

Oh man, those facial images are deep in the uncanny valley. Maybe on the rising side of the slope now, but still way down there.

I was impressed with the Labeled Faces in the Wild (LFW) facial auto-completion results, especially since the system was not trained on LFW at all! The results seemed almost too good to be true. Perhaps this is a testament that there isn&#x27;t <i>that</i> much diversity in human faces?<p>Very well written overall, and I appreciated the author&#x27;s thoughts on TensorFlow+torch at the end of the article.<p>Adversarial training is a fascinating idea, and I love the sound of it. I&#x27;d like to start applying that concept in the future.

On a similar vein...<p><a href="http:&#x2F;&#x2F;www.pyimagesearch.com&#x2F;2016&#x2F;08&#x2F;10&#x2F;imagenet-classification-with-python-and-keras&#x2F;" rel="nofollow">http:&#x2F;&#x2F;www.pyimagesearch.com&#x2F;2016&#x2F;08&#x2F;10&#x2F;imagenet-classificat...</a><p>Uses TF deep learning to classify an image.

Could this be used to generate unique images from a training set?