GitHub - researchmm/STTR: [ACCV'22] Fine-Grained Image Style Transfer with Visual Transformers

Install

conda install -c pytorch pytorch torchvision
conda install cython scipy

Download pre-trained model here

Put it under ./checkpoint_model

Please change the images you want to use in the following folders:

inputs/content: a folder to store all content images
inputs/style: a folder to store all style images

Then run

python demo_sttr_image.py

You could see results in outputs/test_outputs/0005

Name	Name	Last commit message	Last commit date
Latest commit tlatlbtle init Dec 6, 2022 ff52ccf · Dec 6, 2022 History 4 Commits
datasets	datasets	init	Dec 6, 2022
inputs	inputs	init	Dec 6, 2022
models_istt	models_istt	init	Dec 6, 2022
util	util	init	Dec 6, 2022
.gitignore	.gitignore	Initial commit	Oct 4, 2022
LICENSE	LICENSE	Initial commit	Oct 4, 2022
README.md	README.md	init	Dec 6, 2022
demo_sttr_image_ACCV.py	demo_sttr_image_ACCV.py	init	Dec 6, 2022
engine.py	engine.py	init	Dec 6, 2022
hubconf.py	hubconf.py	init	Dec 6, 2022
main_style_transfer.py	main_style_transfer.py	init	Dec 6, 2022