The Picard Maneuver@lemmy.world to Funny@sh.itjust.works · 1 year agoIt's so overlemmy.worldimagemessage-square141fedilinkarrow-up11.12K
arrow-up11.12KimageIt's so overlemmy.worldThe Picard Maneuver@lemmy.world to Funny@sh.itjust.works · 1 year agomessage-square141fedilink
minus-squareAnticorp@lemmy.worldlinkfedilinkEnglisharrow-up5·1 year agoNow it’s making me identify developed pictures from a photo negative. I’m not quite sure what they’re going to do with that training since computers can already perform that task.
minus-squareTheOakTree@lemm.eelinkfedilinkarrow-up3·1 year agoAlso the “select the image below containing the example image above.” Like… we already have computers that can recognize image repetitions.
minus-squareCethin@lemmy.ziplinkfedilinkEnglisharrow-up5·1 year agoSo that’s almost certainly trying to gather data to defeat data poisoning. The other image is probably slightly altered in a way you can’t detect.
minus-squarebitwolflinkfedilinkarrow-up2·1 year agoA common OCR tactic is to turn the image negative and bump the contrast to make text easier to recognize. It could be a precursor for that step.
Now it’s making me identify developed pictures from a photo negative. I’m not quite sure what they’re going to do with that training since computers can already perform that task.
Also the “select the image below containing the example image above.”
Like… we already have computers that can recognize image repetitions.
So that’s almost certainly trying to gather data to defeat data poisoning. The other image is probably slightly altered in a way you can’t detect.
A common OCR tactic is to turn the image negative and bump the contrast to make text easier to recognize.
It could be a precursor for that step.