How to Tag Race Photos When AI Misses the BIB
Bib tagging software should not stop at the photos where the number is easy to read. Hands, medals, jackets, and blur hide bibs on the shots participants often want most. After AI processing, organisers can tag those photos by hand so they still appear in bib search.
TL;DR: Use 9Pic BibTrack first. Then open Photos → Tools → Manual Tagging and add the bibs the pipeline could not read. Pair that with FaceFind so a hidden bib does not hide the person. This is not a tagging vendor. It is exception handling after AI.
Why race photo tagging still needs a human pass
Photographers who search for bib tagging software are not looking for a desktop album organiser. They need a way to attach race numbers to thousands of finish-line and course photos so participants can find themselves. 9Pic BibTrack reads visible bibs automatically. That is the right first step.
It is not the last step. The photos people share most are often the ones where the bib is hardest to read: a medal covering the number, a jacket zipped over the vest, a bottle in the way, motion blur at the chute. A tagging vendor hired to key every file still cannot invent a number that is not in the frame. The practical workflow is hybrid: let AI tag the readable majority, then add the exceptions by hand.
That is different from sending a gallery to a manual race photo tagging service. The dashboard tool is for the photos already in your event, after processing, when you can see which shots came back without a bib.
What manual bib tagging does
Manual Tagging lives under Photos → Tools. It shows processed photos in a gallery. You look up a shot, open it, enter the correct bib number or numbers, and save. Those tags merge with whatever BibTrack already detected. The photo becomes searchable under the bibs you added.
AI tags the readable set
BibTrack handles digit count, ranges, and alphanumeric patterns on photos where the number is in view.
You fill the misses
Open the photo, add the bib, save. The tag is for search. It does not wipe detections that already exist.
FaceFind covers the rest
If the bib never appears, selfie search still finds the person. Manual tagging is for identifiable numbers, not guesswork.
When to tag by hand, and when not to
Manual tagging pays off after you have processed a batch and can review the gaps. It is the wrong first move on a 20,000-photo race. Tune detection first: digit length, category ranges, and pattern rules in Photo Configuration. Re-process if those rules changed after import. Then tag the leftovers.
- Do tag finish and medal photos where you can still read the number even if the pipeline missed it.
- Do tag group shots with two or three visible bibs that should all be searchable.
- Do not tag photos where no number is visible. Use FaceFind instead of inventing a bib.
- Do not send the whole gallery to a vendor to key IPTC fields by hand. That is slow, expensive, and still fails on covered bibs.
How to tag missed bibs in the dashboard
- Process the event photos so BibTrack and FaceFind have run.
- Open the event, go to Photos → Tools, and choose Manual Tagging.
- Search or browse to a photo that should be findable by bib but is not.
- Enter the bib number or numbers and save. The photo joins bib search for those values.
- If many similar misses remain, widen digit or range rules in Configuration before the next batch.
Keep name search and sponsor search on the same event if they matter. A tagged bib still needs a participant list if families will search by name, and a keyword mapping if you want branded moments findable.
Where this sits next to BibTrack and FaceFind
| Job | Use | When |
|---|---|---|
| Read visible bibs at volume | BibTrack | During processing, after you set digit, range, or pattern rules. |
| Fix a missed readable bib | Manual Tagging | After processing, on specific photos. |
| Find the person when the bib is gone | FaceFind | Always keep selfie search on for races. Hidden bibs are common. |
| Let family search by name | Name search | After you have a bib-to-name CSV from registration or timing. |
Race photographers comparing tagging tools already land on peers such as Tagily, SportPxl, RaceTagger, and FlashFrame. The question that matters is not “can software read a clean bib?” It is whether the gallery still works when the bib is not clean. Hybrid tagging plus selfie search is the honest answer.
FAQ
Does manual tagging replace BibTrack?
No. Manual tagging fills gaps after AI. Process first, then tag the photos where the number was readable to a person but not to the pipeline.
Will a manual tag overwrite the AI bib?
No. You add bibs so the photo is searchable under them. The tool is built to complete coverage, not to wipe detections.
Should I hire a race photo tagging service instead?
Only if you have a small set of exceptions and no in-house reviewer. For volume races, a vendor that keys every file is slow and still fails on covered bibs. See the tagging-service comparison for that trade-off.
What if I cannot see the bib at all?
Do not guess. Turn on FaceFind so the participant can find the photo with a selfie. Manual tagging is for numbers you can confirm.
Next steps
Process the gallery with BibTrack, keep FaceFind on, then spend a short review pass on the photos that still have no bib. That is how race photo tagging stays accurate without sending the whole event to a typing queue. Book a demo if you want this mapped onto your next race.