search

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

Super Resolution Demo #13

Closed nelaturuharsha closed 4 years ago

nelaturuharsha commented 4 years ago

Hello, Thank you so much for the contribution. I read in a closed issue you planned to release a Super Resolution Demo, would you be releasing the same?

If possible, could you please nudge me in the right direction w.r.t to implementing it?

Thank you!

aimerykong commented 4 years ago

Hello,

Thanks for your interest! Unfortunately I don't have the mood to release super-resolution demo for now, as I'm busy with our paper re-submission. Only after we get our paper accepted, I'll come back to release:-)

Based upon the released deblurring demo, I think the tweak for super-resolution is to write your dataset function. For example, instead of blurring the patches, you can consider down-sample and up-sampling the image, then crop the corresponding patches to train the model (treat the pairs as input and ground-truth).

Hope this helps!

Regards, Shu

On Wed, Mar 11, 2020 at 4:31 AM Sree Harsha Nelaturu < notifications@github.com> wrote:

Hello, Thank you so much for the contribution. I read in a closed issue you planned to release a Super Resolution Demo, would you be releasing the same?

If possible, could you please nudge me in the right direction w.r.t to implementing it?

Thank you!

— 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/13?email_source=notifications&email_token=ABRJSJHCCTSTTYGIKD3KG33RG5D43A5CNFSM4LFQJW32YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IUEG3ZA, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABRJSJHQG4HLIN35JZNLFZDRG5D43ANCNFSM4LFQJW3Q .