search

yikaiw / CEN

[TPAMI 2023, NeurIPS 2020] Code release for "Deep Multimodal Fusion by Channel Exchanging"
MIT License
288 stars 44 forks source link

some question for image size #7

Closed jianyulei1 closed 3 years ago

jianyulei1 commented 3 years ago

Thanks for your excellent work！I want to input images with different height and width, but I will get an error.Do the images input have to be the same height and width? Thank you very much and look forward your reply.

yikaiw commented 3 years ago

Hi, thanks for your interest. You can change the height and width as long as you adjust SHORTER_SIDE and CROP_SIZE in the config.py file. Could you provide the error so that I can further check it?

jianyulei1 commented 3 years ago

Thank you for your reply, I use image2image_translation, but the default input is 256 256. I want to input a 640 480 image, how do I adjust the code?

yikaiw commented 3 years ago

For image2image_translation, it may be not easy to change the size, since the generated images may be confused. There are works that deal with high-resolution image2image translation.

jianyulei1 commented 3 years ago

您好，您之前给我回复说有处理高分辨的image2image的代码，您能提供给我吗，我对这个很感兴趣想学习一下，非常感谢您的回复。

---Original--- From: "Yikai @.> Date: Wed, Jun 2, 2021 00:19 AM To: @.>; Cc: @.**@.>; Subject: Re: [yikaiw/CEN] some question for image size (#7)

For image2image_translation, it may be not easy to change the size, since the generated images may be confused. There are works that deal with high-resolution image2image translation.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

yikaiw commented 3 years ago

A typical work for high-resolution image2image translation is https://github.com/NVIDIA/pix2pixHD, though it is not multimodal-supported.

jianyulei1 commented 3 years ago

非常感谢您的回复，祝您工作顺利。

---Original--- From: "Yikai @.> Date: Wed, Jun 2, 2021 00:37 AM To: @.>; Cc: @.**@.>; Subject: Re: [yikaiw/CEN] some question for image size (#7)

A typical work for high-resolution image2image translation is https://github.com/NVIDIA/pix2pixHD, though it is not multimodal-supported.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.