Old Photo Restoration via Deep Latent Space Translation (https://paperswithcode.com/paper/old-photo-restoration-via-deep-latent-space)

I am part of a graduated semester project. Do you know if we can find a photo database related to this article? We would be interested in reproducing their work via programming. We need at least 1000 pictures for the training phase.

So far, we have already asked to provide us the data.

  • In these cases it's best to first request data from the authors – philshem Oct 14 '20 at 15:32
  • I have already done that. – David Oct 14 '20 at 15:49
  • all of these things should be added to the text of your question. – philshem Oct 14 '20 at 15:51

There is a github repository linked: https://github.com/microsoft/Bringing-Old-Photos-Back-to-Life

enter image description here

And in that repository there are folders for test-images and images



  • Yeah, but there is not enough pictures there. I can barely see 20 or 30 pictures. We need at least 1000 or even more pictures. – David Oct 14 '20 at 11:15
  • please update your question – philshem Oct 14 '20 at 11:56
  • Ok, I will do that – David Oct 14 '20 at 12:47

Unfortunately, there aren't a lot of relevant data to photo restoration, but imo you can replicate them/ generate your own. I suggest choosing a starting data-set similar to your target data, so if you are trying to restore portrait images maybe the IMDB-WIKI – 500k+ face images with age and gender labels is a good choice. Using OpenCV you can convert these images to black and white (gray scaled) and then you can merge/ blend these images with different old pictures textures. For that see Adding (blending) two images using OpenCV .

With some further tweaking (some randomized degradation and augmentation) the resulting images should be quite realistic and close to real data.

Alternatively, you can use restoration tools/ codes like the Algorithmia- API to restore pictures and build a dataset but that will not result in a better restoration neural network than theirs.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.