![](https://lemmy.dbzer0.com/pictrs/image/31a557ac-0d6d-4d3d-a637-29caa4e15366.png)
![](https://beehaw.org/pictrs/image/1be75b15-2f18-429d-acf7-dcea8e512a4b.png)
It’s too bad that GLIMPSE fork never took off.
It’s too bad that GLIMPSE fork never took off.
They’ve been working on porting it since back in 2012, and didn’t want to redo a bunch of the porting work before they even released it.
I’ve had it freeze up on me several times, where I had to reset the app to get it working again. It works most of the time, but I wouldn’t recommend it yet for general use.
With AI upscaling it fills it in based on the training from other images/videos. So it probably won’t be an alien, but small details common in other videos that looked similar will also show up in the upscaled videos. If an extra flower shows up in a field of grass it’s usually not a big deal, but for some things like faces or symbols, small details can really change the way people interpret it.
Depending on the context it’s probably not that bad, but there’s plenty of details in youtube videos that people pay attention to, like in news, history, tutorials, educational content, and so on. Even for a fictional story, it could add nonsense that people assume is part of the actual show.
The problem with AI upscaling is that it does add something. It fills in the details with things that could plausibly be there, regardless of if they are. It’s especially dangerous if it’s used for something like security footage, where it’ll do stuff like make up a face based on a few pixels.
None of the options are that great right now. I still mostly use Jerboa, and go to the PWA for things that don’t work. Other android apps you can check out are thunder that has an alpha out, memmy which only has an iOS beta so far, but plans to be cross-platform, and lemmur which is outdated and is incompatible with current lemmy, although there’s a fork with more recent development so it could possibly come back at some point.
The biggest thing is probably non-destructive editing, so you can do stuff like apply filters without them changing the underlying image. Gtk3 should add better support for tablets and wayland. There’s also better layer tools and font support. A lot of it was on the backend, which should eventually allow for using other color spaces like cmyk natively.