search

zfu006 / TAP

Official PyTorch implementation of "Temporal As a Plugin: Unsupervised Video Denoising with Pre-Trained Image Denoisers" in ECCV 2024.
16 stars 1 forks source link

Inquiry on Applying TAP for Denoising Real Images #1

Open tiffany-yang opened 4 days ago

tiffany-yang commented 4 days ago

Dear Author, Thank you for your excellent work! I am interested in applying your denoising algorithm to my own real images. Could you please provide guidance on how to achieve this?

zfu006 commented 4 days ago

Do you want to apply TAP to your own images or videos? In raw domain or sRGB domain

tiffany-yang commented 3 days ago

Thank you for your prompt response. I appreciate the detailed information you provided. However, I would like to clarify that my intention is to utilize the method within the RGB domain, rather than the sRGB or RAW domains. Could you please advise if this is feasible, and if so, provide any necessary adjustments or considerations that should be taken into account?

zfu006 commented 1 day ago

I see, I'll upload the code and pretrained model for sRGB domain, but the pretrained model can only deal with synthetic gaussian noise in sRGB domain. If you want to deal with sRGB real noise, you need to retrain the model with your own datasets