I route my dialogue tracks to a mono aux 'DNS1000' and then depending on the project I either:
1. Route it back to another mono audio track and re-record, mute and move the original clip and replace it with a re-recorded version.
2. Route it to my DX bus which goes straight to the master aux and gets re-recorded 'live' as both a DX stem and part of the full mix.
Like Charles, I don't take photos (I used to, but it's not really worth it) - it takes less time to manually re-adjust the faders rather than find that specific pic on your phone.
Re-recording and moving clips isn't the most elegant way of working but I can live with that until I get a DNS One.
FIY - I've asked a similar question here a few months back: Hardware Cedar DNS workflow in surround projects