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
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 |