aimerykong / predictive-filter-flow

Predictive Filter Flow for fully/self-supervised learning on various vision tasks
https://www.ics.uci.edu/~skong2/mgpff.html
140 stars 24 forks source link

Will you provide the code and model for JPEG compression artifact reduction? #14

Open zhengxinChenee opened 4 years ago

zhengxinChenee commented 4 years ago

Thanks for sharing amazing awesome work! According to your paper's setting and code (task01_deblur.ipynb), we try to reproduce the results of JPEG compression artifact reduction by retraining the model you provided (epoch-445.paramOnly). However, the training process of the model is unstable, and the performance is not as good as your paper. Will you provide the code and model for JPEG compression artifact reduction? Hope to get your help.

aimerykong commented 4 years ago

Thanks for your interest!

The released model is for image deblurring. It is not for image denoising (i.e., removing JPEG artifacts). So you may not want to fine-tune it. I'd suggest you train the PFF on the jpeg images (and uncompressed counterparts) from scratch.

I will not release the code for now, as I have been trapped in other projects that prevent me from cleaning up my codebase.

Regards, Shu

On Thu, May 28, 2020 at 6:57 AM zhengxinChenee notifications@github.com wrote:

Thanks for sharing amazing awesome work! According to your paper's setting and code (task01_deblur.ipynb), we try to reproduce the results of JPEG compression artifact reduction by retraining the model you provided (epoch-445.paramOnly). The training process of the model is unstable, and the performance is not as good as your paper. Will you provide the code and model for JPEG compression artifact reduction? Hope to get your help.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/aimerykong/predictive-filter-flow/issues/14, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABRJSJAQCERCI4IE3P3FYTDRTY7QXANCNFSM4NM7HNUA .