CaptionID
MatchID finds the athlete. CaptionID names them. See MatchID →
Bind a roster to the folder or the gallery. Type a short code, or pick the athlete from the roster on screen, and one press writes their name and the full caption line to each field you mapped. No reference headshots, no indexing, nothing to switch on first. It runs in both products.
14 days free, card required. Included in both products.
The same fields get written either way, and the three can be used on the same photo.
01 · CaptionID
Type the delimiter, the code, the delimiter again, and it expands into the athlete and the full caption line. Codes work in any metadata field.
Best when you already think in codes and you are fast at it.
02 · CaptionID
The roster bound to the folder sits under the photo. Click the athlete, or press the key assigned to them, and the name and caption line land in each mapped field at once. Nothing to memorise, no code file to keep.
Best when the roster turned over this week, or the team is not one you know by heart.
03 · MatchID
Index reference headshots and facial recognition starts naming on its own, reading jersey numbers when the face is not legible. Anything under your threshold waits for one click rather than being guessed at.
Best when a single game produced hundreds of frames and the deadline is close.
Which is why the roster path is the default one and not the fallback.
Facial recognition has to be shown who people are before it can find them, and being shown means indexing reference headshots for a team. At the start of a season the roster is a spreadsheet somebody already maintains, and the headshots are a session nobody has scheduled yet.
CaptionID identifies nobody. You say who it is, and it writes the name and the caption line. That is a smaller promise, and it is the reason it works on the first gallery rather than the fourth.
Index headshots when you have them and MatchID starts naming ahead of you. The two write the same fields on the same photo, so turning it on changes how a name arrives and nothing about where it lands. Anything MatchID does not reach is one press from the roster, which is CaptionID again.
Neither path replaces the other. One of them just needs less from you first.
Pick a saved roster, upload your own, or build one from the Roster Library.
Column A is the short code, and it has to be unique, because that is the part you type. If two rows carry the same code the first match wins. Everything after column A is yours: name, number, position, class, team, whatever the row needs. You map those columns to the metadata fields you want them in, once, and the mapping holds.
You do not have to build one from nothing. The Roster Library ships with the product and covers NCAA Division I, II and III and the major US pro leagues.
What the Roster Library covers →Type the delimiter, the code, then the delimiter again. Column B comes out by default. Add #2 before the closing delimiter for column C, #3 for column D, and so on. The delimiter is set in Settings and defaults to a backslash.
A token is a placeholder you drop into a metadata field once. Person Name inserts whoever you tag. The long-form tokens build the wire caption styles from the same tagged row, and a part the row does not carry drops out cleanly instead of leaving a gap.
| Token | What lands in the field |
|---|---|
| {person_name} | Marcus Cole |
| {long1} | Riverdale Wolves Marcus Cole (12) |
| {long2} | Marcus Cole #12 of the Riverdale Wolves |
| {long3} | Riverdale Wolves quarterback Marcus Cole |
| {long4} | Riverdale Wolves quarterback Marcus Cole (12) |
Tag two people and the names are joined with “and”. Tag three or more and they run as a list, commas throughout and “and” before the last. You can also select a run of frames and put the same people across all of them in one pass, with per-photo exceptions where a frame differs.
GalleryID Desktop reads existing Photo Mechanic code replacement values as they are. A code set you have maintained for years still expands, and the muscle memory transfers on day one rather than after a migration.
Desktop runs alongside Photo Mechanic and reads its values. It can replace it, and it does not have to. How the two compare →
What differs is where the work happens, which is the difference between the two products generally.
Rosters bind to a folder and subfolders inherit it. Short codes, the on-screen roster and the key assignments all live in Browse, next to ingest, culling and delivery, so naming is part of the pass you were already making rather than a second trip. Nothing is uploaded and nothing is stored on GalleryID servers.
GalleryID Desktop →Rosters live in your account by team and by season, and CaptionID names athletes in an uploaded gallery with a button press. The names then make the archive searchable by athlete, which is what the hosted product is for.
GalleryID Cloud →A name is written one of two ways: somebody confirms it, or it clears a confidence threshold that person set.
On the CaptionID path there is no second route to worry about. A person picked the athlete off the roster, so the name in the file is the name that person chose. There is no score, no threshold and no queue, because nothing was inferred.
That is worth saying plainly because the rest of this category sells the inference. The inference is genuinely useful, which is why MatchID exists. It is not the only way to get a name onto a photograph, and it should not be the price of entry.
Standard IPTC and XMP, not a format of ours.
Names write to the file in XMP, with the legacy IPTC-IIM mirror alongside for the agency tooling and wire services that still read IIM on JPEGs, across the fields you mapped including Persons Shown and Personality. The names go wherever the file goes, so they read in Photo Mechanic, Lightroom, Capture One and Photoshop with no plugin and no re-tagging at the far end.
GalleryID Desktop writes as you tag, before the file leaves your machine. GalleryID Cloud writes on download, and the metadata updates whenever the names change.
Sports photo captioning, in full →It is the roster half of naming a photo. Bind a roster to a folder or a gallery, then name an athlete either by typing a short code or by picking them from the roster on screen. One press writes the athlete name and the full caption line to each metadata field you mapped. It runs in both GalleryID products and it needs no reference headshots, so it is the path that works on day one.
MatchID finds the athlete. CaptionID names them. MatchID is facial recognition and jersey number reading, so it has to be shown who people are first, by indexing reference headshots. CaptionID starts from a roster you already keep and identifies nobody: you say who it is, and it writes the name and the caption line. They are not alternatives. Most people run CaptionID from the start and switch MatchID on later, and the two write the same fields on the same photo.
No. A roster is enough on its own. Indexing headshots is what switches MatchID on, and that is a choice you make later rather than a setup cost you pay before you can name anything. This matters most at the start of a season, when the roster exists and the headshots do not.
A spreadsheet of names. Column A is the short code and it has to be unique, because that is what you type; if two rows share a code the first match wins. The remaining columns are yours, and the app maps them to the metadata fields you choose. You can pick a saved roster, upload your own file, or build one from the GalleryID Roster Library, which ships with NCAA Division I, II and III and the major US pro leagues.
Type the delimiter, the code, then the delimiter again, and it expands. Column B comes out by default. Add #2 before the closing delimiter to pull column C instead, #3 for column D, and so on, so one code reaches whichever part of the row you need. The delimiter is set in Settings and defaults to a backslash. Codes work in any metadata field, not just the caption.
Yes. GalleryID Desktop reads existing Photo Mechanic code replacement values as they are, so a code set you have maintained for years keeps working and the muscle memory transfers on day one. Desktop runs alongside Photo Mechanic and reads its values; it can replace it and it does not have to.
Each name lands in the fields you mapped, and the names join the way a person would write them: two are joined with "and", three or more use commas with "and" before the last. The long-form tokens build wire-service caption styles from the tagged row, lines like "Riverdale Wolves quarterback Marcus Cole (12)", and a missing part of the row drops out cleanly rather than leaving a gap.
Yes, and this is where starting from the roster is worth the most. CaptionID names people from a list rather than from a face, so a covered face changes nothing about it. That is also why the two paths are worth keeping straight: MatchID has jersey number reading for exactly this problem, and CaptionID does not have the problem.
Names write to the file in XMP, with the legacy IPTC-IIM mirror alongside, across the fields you mapped including Persons Shown and Personality. The names go wherever the file goes, so they read in Photo Mechanic, Lightroom, Capture One or anything else that reads photo metadata. GalleryID Desktop writes as you tag, on your own machine. GalleryID Cloud writes on download, and the metadata updates whenever the names change.
No. It is part of both products at no additional charge, the same way MatchID is. GalleryID Cloud is priced on storage alone, from $20/mo for 200 GB. GalleryID Desktop is $20/mo or $200/yr and the subscription covers two computers. Both include a 14-day free trial and both require a card.
Yes. A roster is a spreadsheet of names, so a start list, a delegate list, a cast sheet or a class list binds to a folder exactly the same way and one press writes the name and the caption line. Only two features anywhere in the product are sports-specific, and both of them are inside MatchID.
You already have the roster.Start there.
CaptionID is in both products at no extra charge. GalleryID Cloud is priced on storage alone, from $20/mo for 200 GB. GalleryID Desktop is $20/mo or $200/yr and covers two computers.
14 days free, card required. No free plan on either product.