Hello world. For those who are new to my channel im a big fan of deep learning {who isn't}. I've made about a BILLION deep dreams. I've been an artist for many years: But im also intensly interested in what A.I can create artistically. Especially A.I assisted artwork where you have control over the inputs.
Remember when this site was really popular that made peoples faces that didn't really exist? { https://thispersondoesnotexist.com/ } Okay ... to be fair I made this generation randomly and there are some weird anomalies on the right side of the face. However, I made some landscapes today that look very real and they do not exist in real life.
There are probably small anomalies in this landscape that I created with A.I However, without zooming in: This photograph looks 100% like it exists in real life. It looks like a real location! I love the intense contrast. It reminds me of when I used to cut strips of photo prints from 35mm and glue them together very carefully. This created unique landscapes that don't really exist.
Here is a simulated photo post sunset with a bit more light coming through the "camera". I think this deep fake is pretty damn convincing! I managed to line up the sealine even more perfect the second time. I used an obscure tool called http://nvidia-research-mingyuliu.com/gaugan/that I learned about from Hive account@kamchore . This tool is used to draw digital landscapes. I used it like a deep learning tool and trust me its MUCH faster than deep learning and theres no login. Should I try a non-landscape picture like a face and an oil painting?
. This tool is used to draw digital landscapes. I used it like a deep learning tool and trust me its MUCH faster than deep learning and theres no login. Should I try a non-landscape picture like a face and an oil painting?
