search

CompVis / adaptive-style-transfer

source code for the ECCV18 paper A Style-Aware Content Loss for Real-time HD Style Transfer
https://compvis.github.io/adaptive-style-transfer/
GNU General Public License v3.0
723 stars 139 forks source link

Any plan for releasing the pre-trained model for calculating style transfer deception rate? #5

Closed ycjing closed 5 years ago

ycjing commented 5 years ago

Hi. Thanks for the nice code. The results are really impressive. It would be great if you could release the pre-trained model for calculating style transfer deception rate and follow-up works could thus have a quantitative comparison with your algorithm. Thanks!

asanakoy commented 5 years ago

Hi, Yes, we will try to release it next week.

пт, 21 сен 2018 г., 14:52 Yongcheng Jing notifications@github.com:

Hi. Thanks for the nice code. The results are really impressive. It could be great if you could release the pre-trained model for calculating style transfer deception rate and follow-up works could thus have a quantitative comparison with your algorithm. Thanks!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/CompVis/adaptive-style-transfer/issues/5, or mute the thread https://github.com/notifications/unsubscribe-auth/ABnM9Va78Ijut9oeotrNxlMUXqoC8N3dks5udNLlgaJpZM4Wz_4W .

Shihab-Shahriar commented 5 years ago

I need the same model, but for style grouping mainly. @asanakoy, it'd be great help to have it sooner. If not possible, do you know any other publicly available model that did the same?

Thanks.

asanakoy commented 5 years ago

The instructions how to calculate deception score are now in evaluation folder.

Run evaluation/download_evaluation_data.py to download a pretrained model for deception rate computation or get it by the link https://yadi.sk/d/A2CBqSGuJ0M_XA.

I'm sorry for the delay.