Dear Reader:

You are viewing a story from GN Version 5.0. Time may not have been kind to formatting, integrity of links, images, information, etc.

Creating photorealistic images with neural networks and a Gameboy Camera

by rawmeatcowboy
21 February 2017
GN Version 5.0

Back in 1998, the Gameboy camera got the world record as “smallest digital camera” in the Guinness book of records. An accessory you could buy was the small printer you could use to print your images. When I was 10 years old we had one of these cameras at home and used it a lot. Although we did not have the printer, taking pictures, editing them, and playing minigames was a lot of fun. Unfortunately, I could not find my old camera (no colored young Roland pictures, unfortunately), but I did buy a new one so I could test my application.

In the end the result turned out very good. The generated images are really great. Although we trained on a small part of the face even pictures of whole heads seem to turn out nice. The following images are images the network has never seen before. The image in the center is the “real image”, the image on the right is generated by our neural network. Note that even skincolor is accurate most of the times.

Full experiment here (thanks MarioCheated!)

[Link]