Live data from Hacker News

Unified GAN for Image-to-Image translation

21–22 of 22 posts

Re: Unified GAN for Image-to-Image translation

#21

Post by IshKebab » Thu, Nov 30, 2017, 11:04 AM UTC

amelius wrote: post #19

bambax wrote: post #13

In the examples, the source images are very very uniform (Caucasian, young, healthy). Does this work with other images? Can it age photos of people who are already over 50 for example? Or non-whites? Or children? Etc.

I wish researchers showed a random mix of results, not just their best results ...

Some do - they show failure modes.

Re: Unified GAN for Image-to-Image translation

#22

Post by bitL » Thu, Nov 30, 2017, 11:56 AM UTC

bambax wrote: post #13

In the examples, the source images are very very uniform (Caucasian, young, healthy). Does this work with other images? Can it age photos of people who are already over 50 for example? Or non-whites? Or children? Etc.

I frankly don't see why it couldn't add more years to 50y. Moreover, it could likely "unage" over 50y to 20y, just by inverse of their aging method. GANs are unbelievably powerful.

Post reply on HN