Comment on Court Bans Use of 'AI-Enhanced' Video Evidence Because That's Not How AI Works
ChaoticNeutralCzech@lemmy.one 5 months ago
Sure, no algorithm is able to extract any more information from a single photo. But how about combining detail caught in multiple frames of video? Some phones already do this kind of thing, getting multiple samples for highly zoomed photos thanks to camera shake.
Still, the problem remains that the results from a cherry-picked algorithm or outright hand-crafted pics maybe presented.
Natanael@slrpnk.net 5 months ago
Depends on implementation, if done properly and if they don’t try to upscale and deblur too much then that kind of interpolation between multiple frames can be useful to extract more detail. If it’s a moving subject then this type of zoom can create false results because the algorithm can’t tell the difference and will think it’s an optic artifact. For stationary subjects and photographers it can be useful