apftwb@lemmy.world to Technology@lemmy.worldEnglish · 23 hours agoRecreating uncensored Epstein PDFs from raw encoded attachmentsneosmart.netexternal-linkmessage-square56fedilinkarrow-up1585arrow-down15
arrow-up1580arrow-down1external-linkRecreating uncensored Epstein PDFs from raw encoded attachmentsneosmart.netapftwb@lemmy.world to Technology@lemmy.worldEnglish · 23 hours agomessage-square56fedilink
minus-squareTaldan@lemmy.worldlinkfedilinkEnglisharrow-up3·8 hours agoOCR is mostly good enough. Problem here is we have 76 pages that we need to be read perfectly, with a low fidelity input We also have very little in the way of error correction, since it’s mostly not human readable
OCR is mostly good enough. Problem here is we have 76 pages that we need to be read perfectly, with a low fidelity input
We also have very little in the way of error correction, since it’s mostly not human readable