Simulating the GAN Network via Python, and understanding how it works.
Brian Nguyen Alex Chuoy Danny Chu
While we are unable to successfully implement a complete AI in order to successfully pull of GAN, via the PILLOW module, we use a series of brute forced function to simulate an AI creating and attempting to distinguish if the image is real or not. Most of our functions are lifted from the link below.
To run this program, you must download the PILLOW module in order to run it, any images will work, just label them appropriately as stated in the code, the random module, and lastly be up to date with python.
Works Cited:
Alex Clark and Lundh, Fredrik. “ImageDraw Module.” ImageDraw Module - Pillow
(PIL Fork) 3.1.2 Documentation, 2016,
<pillow.readthedocs.io/en/3.1.x/reference/ImageDraw.html>.
Oberhaus, Daniel. “This AI Will Turn Your Dog Into a Cat.” Vice, VICE, 16 Apr. 2018,
<www.vice.com/en_us/article/kzxkvm/neural-network-
transforms-dogs-into-cats-ai-gan.>
“Super Smash Brothers Ultimate - Fight for an Invite.” Know Your Meme, Literally Media,
23 Apr. 2019 <https://knowyourmeme.com/photos/1485209-
super-smash-brothers-ultimate>
“Super Smash Brothers Ultimate - Joker's New Ride.” Know Your Meme, Literally Media,
27 Apr. 2019 <https://knowyourmeme.com/photos/1486866-
super-smash-brothers-ultimate>
“Super Smash Brothers Ultimate - Mario's alt Final Smash.” Know Your Meme, Literally Media,
28 Apr. 2019 <https://knowyourmeme.com/photos/1487158-
super-smash-brothers-ultimate>
“Super Smash Brothers Ultimate - Smash Bros live action movie.” Know Your Meme, Literally Media,
03 May. 2019 <https://knowyourmeme.com/photos/1488778-
super-smash-brothers-ultimate>