zlacker

[return to "RealFill: Image completion using diffusion models"]
1. Jorge1+3a[view] [source] 2023-09-29 19:16:03
>>flavor+(OP)
Wow. The use case that comes to mind for me is when you take a big family photo (or 20) and someone inevitably ends up cut-off by accident.

So then you just feed RealFill the 20 pictures you took and your uncle is magically painted in.

◧◩
2. jetrin+Fa[view] [source] 2023-09-29 19:20:32
>>Jorge1+3a
Also getting everyone smiling with their eyes open at the same time. Phone cameras could record a group photo for five or ten seconds and use the best expression from different times for each person.
◧◩◪
3. patapo+Bb[view] [source] 2023-09-29 19:25:58
>>jetrin+Fa
Or you take a single picture of a group in front of a monument, but cut it off. As I understand it you could find pictures of the monument online, run the model, and have a picture with the group and the entire monument.

Probably google can even do this automatically - I would not be surprised if I get suggestions to fix images with cut off buildings via Google Photos in the future! Would be so cool.

[go to top]