PrivyMark

Response from developer

Thanks so much for this — really glad the detection list and the manual toggles are working for you. One thing you can do right now: word-level redaction already exists, it's just behind a gesture. Long-press any text in the photo and that line fans out into individual word chips — tap only the words you want covered and leave the rest readable. There's a "cover whole line" button in the same sheet if you do want all of it. That said, you're right about the detected items. When you toggle an OCR result, it applies the box the detector produced, and for some types — addresses especially — that box is the whole line. That's a real gap, and it's being fixed: the next update adds a refine button on each detected item that opens the same word picker, pre-filled with what was detected, so you can trim it down word by word. Detection boxes are being tightened too, so they stop bleeding into neighbouring words. Thanks for taking the time to write this up — it's exactly the kind of feedback that makes the app better.