Full STEEM Ahead - original STEEM art (and BONUS alien tentacle monster)

EDIT: CC Commons Attribution-Share Alike license added

This image is licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license

This is the license the original train image was released with on WikiCommons so all derivative works, including this and anything you do with my derivative work above, must be released under the same license with attribution to the original author.

I am linking to the original Wiki Commons page so please follow that link for full attribution and I release the derivative images contained in this post under the same license, meaning you can use them however you want as long as you attribute the original Wiki Commons author as well as myself and you release your new work under the same license.

/EDIT

This has gone through many iterations since I posted a work in progress 4 months ago. This is mixed media, PhotoShop paint on top of a photo of a locomotive from WikiCommons and the result ran through the Deep Dream Generator before additional PhotoShop painting and image manipulation on top of the resulting image.

For reference here is the 4 month old version that I originally posted:

I spent a little while PhotoShop painting on top of a higher res version of the above to fill in the blank spots, but ended up scrapping that for a different technique. I painted on the original image of the train to add details like smoke turning into the Steem logo shapes, and I painted over landscape background as well to give the AI more to sink its teeth into. Here is the image I fed into the neural network:

Here is the style image I used (this is not my own and is shown here as fair use in reporting my process - the neural network applies the style of this image to the base image I uploaded, but does not actually copy the image:

Here is the raw output from the Deep Dream Generator before the final stages of PhotoShop editing:

Bonus Tentacle Monster


Art by moi

H2
H3
H4
Upload from PC
Video gallery
3 columns
2 columns
1 column
14 Comments