Some models trained by different people using the chainer fast neuralstyle app by yusuketomoto. Please note, there has been a change on how the models are trained which makes them not suitable for the most recent version relased. Will update here soon.
The animated gifs have been created with a simple "real-time" video capture wrapper written in python. The code is available at https://github.com/gafr/chainer-fast-neuralstyle-video
Compatible with old code (0d96350)
| ModelName | Reference | Parameters | Style | Samples |
|---|---|---|---|---|
| starrynight | chainer-fast-neuralstyle | unknown |  |
  |
| cubist | cubistMirror | unknown |  |
  |
| hokusai | cubistMirror | unknown |  |
 |
| hundertwasser | - | 1000 training images, 2 epoches, training: size 512 (full) |  |
 |
| kandinsky | uploaded by 6o6o | full coco training set, 2 epoches, training size 512 (cropped) |  |
  |
| kandinsky | uploaded by 6o6o | full coco training set, 2 epoches, training size 512 (full) | |
 |
| edtaonisl | uploaded by 6o6o | full coco training set, 2 epoches, training size 512 (cropped) |  |
  |
| Model Name | Reference | Parameters | Style | Samples |
|---|---|---|---|---|
| Kanagawa | 6o6o | full coco, 2 epochs, image_size 512 (crop) |  |
 |
| Candy | Heartsie | full coco, 2.6 epochs, image_size 512 (crop) |  |
 |
| Fur | 6o6o | full coco, 1 epoch, image_size 512 (crop) |  |
 |
| Starry Night | DylanAlloy | full coco, 2 epochs, image_size 512 (crop) |  |
 |
| The Scream | DylanAlloy | full coco, 2 epochs |  |
 |