init

README.md

@@ -1,2 +1,28 @@
 # ArcaneGAN
 ArcaneGAN
+ArcaneGAN v0.1 by Alex Spirin
+This is a proof of concept release. The model is in beta (which means it's beta than nothin')
+
+Here are some image pairs. I've specifically picked various images to see how the model performs in the wild, not on aligned and cropped faces.
+
+![0_256_](https://user-images.githubusercontent.com/11751592/144793668-be43ba18-a8c3-4b37-ba95-fc7bf56c5648.jpg)
+![258c27bcb658a86765361c1faca7b749fa3a36aaf07e975b408281c0a9c76513](https://user-images.githubusercontent.com/11751592/144793712-11f7438f-0f14-4a50-bae6-673a1461ba71.jpg)
+![e42](https://user-images.githubusercontent.com/11751592/144793847-121d6c17-f2a2-484d-a2ed-35f53f27197c.jpg)
+![ewewe](https://user-images.githubusercontent.com/11751592/144793886-e3063e9c-fcb7-4570-82cb-0bcc3bc42b58.jpg)
+![maxresdefault](https://user-images.githubusercontent.com/11751592/144793912-c58ff819-1f13-4bf8-89da-643fe08889fa.jpg)
+![photo_2021-11-16_19-32-15](https://user-images.githubusercontent.com/11751592/144793930-e85dcd44-3417-4bff-b170-e6b39a2f57ca.jpg)
+![photo_2021-11-16_19-34-02](https://user-images.githubusercontent.com/11751592/144793943-63bdecf1-c5e0-4b05-8f31-0dd14fd8c124.jpg)
+![photo_2021-11-16_19-34-33](https://user-images.githubusercontent.com/11751592/144793967-bdc73e78-abe9-4678-86cf-7fd71545963f.jpg)
+![photo_2021-11-16_19-34-49](https://user-images.githubusercontent.com/11751592/144793979-c03216a1-414f-41c5-8f46-492d9be09d40.jpg)
+![photo_2021-12-04_07-23-17](https://user-images.githubusercontent.com/11751592/144794058-1404130d-847b-4e3e-b66f-22e859c57666.jpg)
+![photo_2021-12-04_07-48-29](https://user-images.githubusercontent.com/11751592/144794159-19d48ac0-79f2-4cc2-867b-fafb4acb9191.jpg)
+![photo_2021-12-04_08-06-40](https://user-images.githubusercontent.com/11751592/144794243-b64b2cf9-2032-4022-9a34-94d65b283566.jpg)
+![photo_2021-12-04_08-07-04](https://user-images.githubusercontent.com/11751592/144794253-600c0bc4-5176-4142-81d5-e389aaf1e917.jpg)
+![photo_2021-12-04_11-26-27](https://user-images.githubusercontent.com/11751592/144794261-b6f868ca-8dfb-4ea2-9e60-66cf683b2569.jpg)
+
+
+It does something, but not much at the moment. 
+
+The model is a pytroch *.jit of a fastai v1 flavored [u-net](https://fastai1.fast.ai/vision.models.unet.html) trained on a paired dataset, generated via a blended stylegan2. You can see the blending colab I've used [here](https://github.com/Sxela/stylegan3_blending). 
+
+Inference notebook is [here](https://colab.research.google.com/drive/1r1hhciakk5wHaUn1eJk7TP58fV9mjy_W?usp=sharing)