News for hackers. Stuff that matters.

A comment on 3D photographic inpainting from a single source image. Parent comment

Re:3D photographic inpainting from a single source image by fxtentacle on Saturday April 11, 2020 @04:14AM (#22840374)
Problem is, this paper won't be able to produce those results from historical photos. I find them quite misleading, to be honest.

You'll need to either use a separate 3D depth estimation AI, or (more likely) have someone do a manual stereoscopic 3D conversion of your historical image. Only then (when you have depth data) can the algorithm presented in this paper start its work.

[ Parent | on HN ]
Re:3D photographic inpainting from a single source image by johndough on Saturday April 11, 2020 @05:48AM (#22840723)
> You'll need to either use a separate 3D depth estimation AI

They seem to be using MiDaS https://github.com/intel-isl/MiDaS for depth estimation which does a reasonable job on some random image pulled from pixabay https://i.imgur.com/IfbeaqY.jpg

[ Parent | on HN ]
Re:3D photographic inpainting from a single source image by fxtentacle on Saturday April 11, 2020 @06:43AM (#22840969)
That looks good, indeed :)
[ Parent | on HN ]
Re:3D photographic inpainting from a single source image by fxtentacle on Sunday April 12, 2020 @07:13AM (#22847436)
In the meantime, I tried our Midas on other images and it failed most of the time. Really only works for those stock-is pictures with clear foreground and background and I believe it mainly just detects bokeh blur as opposed to actually understanding the scene.
[ Parent | on HN ]
Re:3D photographic inpainting from a single source image by jbhuang0604 on Sunday April 12, 2020 @01:51PM (#22849843)
Yes, you are right. Existing single-image depth estimation models are not far from perfect. We hope to see future development in that direction to further improve the visual quality of 3D photos.
[ Parent | on HN ]