search

microsoft / StyleSwin

[CVPR 2022] StyleSwin: Transformer-based GAN for High-resolution Image Generation
https://arxiv.org/abs/2112.10762
MIT License
510 stars 49 forks source link

Can styleSwin train like pix2pix? #26

Closed AugustLee93 closed 2 years ago

AugustLee93 commented 2 years ago

Hi there, I was wondering, can styleswin train like pix2pix? I wanna use vit to do img2img task, Thanks

ForeverFancy commented 2 years ago

Hi, although we did not provide the modified model for img2img task, I think this is theoretically possible. A simple idea is just replacing the input to AdaIN with image feature like SPADE, you could have a try. Thanks.

AugustLee93 commented 2 years ago

Hi, although we did not provide the modified model for img2img task, I think this is theoretically possible. A simple idea is just replacing the input to AdaIN with image feature like SPADE, you could have a try. Thanks.

Thanks for your replying, I would give a try! Thanks again!