this post was submitted on 07 Apr 2026
1119 points (99.3% liked)
pics
28018 readers
1321 users here now
Rules:
1.. Please mark original photos with [OC] in the title if you're the photographer
2..Pictures containing a politician from any country or planet are prohibited, this is a community voted on rule.
3.. Image must be a photograph, no AI or digital art.
4.. No NSFW/Cosplay/Spam/Trolling images.
5.. Be civil. No racism or bigotry.
Photo of the Week Rule(s):
1.. On Fridays, the most upvoted original, marked [OC], photo posted between Friday and Thursday will be the next week's banner and featured photo.
2.. The weekly photos will be saved for an end of the year run off.
Instance-wide rules always apply. https://mastodon.world/about
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I've been using it as a sort of litmus test for AI images. Even at a high quality setting, AVIF compresses them down to almost nothing.
Something to do with the lack of natural "jitter" in AI images and the way AVIF has been designed to perfectly deal with this.
The diffusion models at least were basically designed to "remove noise from a random image until a real image emerged" so that actually makes a lot of sense, interesting
Oh guess the AI folks haven't yet realized that you can increase perceived accuracy by adding a small amount of random noise.
Which of course they didn't, because then they'd need to understand some things about images instead of just throwing whatever data they can find at it and hoping it figures it all out from that.
Though I'm not sure how much it applies to images, as the examples I've seen were audio. But it's cool to hear a low bitrate audio sample of someone talking unintelligibly and then play that same clip with random noise also playing and suddenly you can understand what is being said.