{"id":434,"date":"2020-07-17T07:38:55","date_gmt":"2020-07-17T07:38:55","guid":{"rendered":"https:\/\/blog.comixify.ai\/?p=434"},"modified":"2020-08-06T11:32:14","modified_gmt":"2020-08-06T11:32:14","slug":"comixify-style-transfer-can-undone-be-done-differently","status":"publish","type":"post","link":"https:\/\/blog.comixify.ai\/index.php\/2020\/07\/17\/comixify-style-transfer-can-undone-be-done-differently\/","title":{"rendered":"Can Undone be done differently?"},"content":{"rendered":"Read-o-Meter<\/span> 3<\/span> minutes<\/span><\/span>\n
\"\"<\/figure><\/div>\n\n\n\n

by  Pawe\u0142 Andruszkiewicz <\/a> <\/p>\n\n\n\n

Last year Amazon Prime released a new animated series called \u201cUndone\u201d. TV series that was entirely created through rotoscope animation. You might be familiar with this technique if you\u2019ve watched movies like \u201cA Scanner Darkly\u201d and \u201cWaking Life.\u201d A similar technique was also used in the \u201cTake On Me\u201d video by A-ha. In \u201cUndone\u201d, all the animation was created by Amsterdam-based animation studio \u201cSubmarine\u201d. A Team of talented painters and artists from all over the World was engaged to create the final, magnificent effect. What is more, the actual rotoscoping, led by co-producer Craig Staggs, was performed by the super experienced team responsible among others for \u201cA Scanner Darkly\u201d. Nevertheless, the whole process is very time consuming and requires lots of manual labour. What if there would be a technology capable of reducing this handwork by 95%, leaving only the hardest and most interesting parts of it to the artist?<\/p>\n\n\n\n

\"\"
Comixify.ai Style Transfer demo vol 2. https:\/\/www.youtube.com\/watch?v=JsZTrHMEH2k<\/a> <\/figcaption><\/figure><\/div>\n\n\n\n\n\n\n\n

Style Transfer – How we do it<\/strong><\/p>\n\n\n\n

It turns out that such technology already exists, it\u2019s called Style Transfer. Firstly introduced a couple of years ago by L. Gatys et al. in the publication called \u201cA Neural Algorithm of Artistic Style\u201d<\/a> Style Transfer has come a long way from a very slow optimization based technique to the most recent methods based on state-of-the-art Machine Learning models. In comixify.ai we developed our own proprietary Style Transfer technology based on Deep Learning Generative Adversarial Networks. Our method is able to clone a style of any artist represented by just a few representative painted or drawn samples and then apply it to any video sequence with the outstanding quality. To demonstrate our capabilities let\u2019s dive in one of our styles that we currently work on. We asked an artist to draw a bunch of examples of style inspired by A-ha \u201cTake On Me\u201d music video. Selected samples can be seen below along with original images.<\/p>\n\n\n\n