Show HN: Meow – An Image File Format I made because PNGs and JPEGs suck for AI • Hacker News Board

Re: Show HN: Meow – An Image File Format I made because PNGs and JPEGs suck for AI

#81

Post by csmantle » Mon, Jun 16, 2025, 2:51 AM UTC

This approach tries to combine pixels stream with metadata stream, but from my opinion that's not a very elegant solution.

Being consistent when perceived and being lossless in information are different things. https://github.com/Kuberwastaken/meow/blob/60339a764a2365c4a... shows that the library simply truncates the lower bits of the pixel, doing a lossy transformation to the carrier. This could lead to (at best) inconsistencies in later sample processing, or (at worst) the sample being pulled away far from the original location in the sample/embedding space.

Steganography is generally used to covertly carry information -- you try to keep your extra bits in-band with the carrier. That's why anti-piracies use that to carry identifiers, watermarks or so, without disrupting the perceived consistency/quality of the carrier.

Metadata, on the other hand, does not need to be transmitted covertly. They are public information can be included in the container format itself. Many image formats already have facilities for these out-of-band data streams. So I think it's reinventing the wheels, but in a crude and complicated way.

Re: Show HN: Meow – An Image File Format I made because PNGs and JPEGs suck for AI

#82

Post by b0a04gl » Mon, Jun 16, 2025, 6:59 AM UTC

mousethatroared wrote: post #61

b0a04gl wrote: post #52

using LSB to store structured metadata inside PNG is clever > survives format conversion, stays invisible to standard viewers, and doesn't break compression. but the space is tight. even at 1 bit per channel, that's just 3 bits per pixel on RGB. given that, how are you handling tradeoffs between spatial fidelity (like masks or edges) versus scalar data (like complexity scores)? is there a priority system, or does it…

3 bits per pixel is in the order of megabyte amounts of information. 640x480 a 3 bits -> 900kB. Thats a lot. Too much. This is why we cant have nice things. Because people keep inventing things to destroy privacy.

any channel that hides data invites misuse, even if the intent is benign. the same technique that enables structure also enables stealth. how easy it becomes to slip data past casual inspection. worth being cautious, even with open use cases

Re: Show HN: Meow – An Image File Format I made because PNGs and JPEGs suck for AI

#83

Post by RamblingCTO » Mon, Jun 16, 2025, 7:37 AM UTC

kuberwastaken wrote: post #37

RamblingCTO wrote: post #5

> it gets stripped way too easily when sharing that's not a bug, that's a (security) feature

Sure is, but do we need that security feature that's common on the internet to have worse context when working with AI?

yes it is. because the risk of forgetting metadata that pwns you is higher than the inconvenience of not having it.

Re: Show HN: Meow – An Image File Format I made because PNGs and JPEGs suck for AI

#84

Post by mousethatroared » Mon, Jun 16, 2025, 11:17 AM UTC

b0a04gl wrote: post #82

mousethatroared wrote: post #61

Earlier quoted context omitted.

3 bits per pixel is in the order of megabyte amounts of information. 640x480 a 3 bits -> 900kB. Thats a lot. Too much. This is why we cant have nice things. Because people keep inventing things to destroy privacy.

any channel that hides data invites misuse, even if the intent is benign. the same technique that enables structure also enables stealth. how easy it becomes to slip data past casual inspection. worth being cautious, even with open use cases

Sure would be nice if folks in your field had ever had a moment of self reflection of the greater societal impact of their work.

Maybe Stallman has had.