Wedding photo face search for photographers and studios
Guests scan a QR code, take a selfie, and see the photos they are in, without scrolling a 3,000-photo gallery.
A wedding gallery is small next to a race or a conference, but the audience is demanding: 150 guests who each want their own photos, and a couple who wants the gallery kept private. The photographer or the gallery platform usually builds this, and the job is to make guest search work without turning the gallery into something anyone with a link can mine.
Who builds this: Wedding photographers, studios shooting many weddings a month, and the gallery platforms that host their work.
the workflow
How it goes together
- 1
One collection per wedding
POST /v1/collectionsName it after the booking, for example `studio-42-wedding-0614`. One wedding's guests never match another's photos, and deleting a couple's data is one call.
- 2
Index the edited gallery, not the raw card dump
POST /v1/batches (op: "index")Upload after culling and editing, so you pay to index the 3,000 photos you deliver rather than the 9,000 you shot. Batch is the lower rate, and nobody is waiting on a live feed.
- 3
Put the selfie box behind the gallery link
POST /v1/collections/{id}/searchGuests reach the selfie box through the couple's private gallery link or a QR code on the tables. One search by image returns ranked photo ids, and you render those photos from your own storage.
- 4
Show strong matches, ask about borderline ones
Each match carries a 0 to 1 similarity score. Show high scores straight away and put lower ones behind an "Is this you?" prompt. Set the cut-offs from your own galleries, not a number copied from another vendor.
- 5
Delete when the delivery window closes
DELETE /v1/collections/{id}Most studios keep a gallery open for a few months. When it closes, delete the collection. A default delete is soft and restorable during a grace window. Pass `immediate=true` when a couple asks for erasure.
Create collection · Batch index · Search · Delete collection
Request and response shapes for every call are in the API reference.
What is different about weddings
- The couple is your client and the guests are not. Guests never agreed to your terms, which is why the selfie box sits behind the couple's link and why the couple should approve turning it on.
- Small galleries, and many of them. A studio pays per wedding, so multiply the example below by your monthly bookings.
- Group shots are what guests want. A table photo with ten guests is one indexing charge and appears in all ten guests' results.
- Removal is per photo. There is no per-face delete: removing a guest means deleting the photos they appear in from the collection, and each photo delete removes every faceprint in that photo.
what it costs
Cost example: One wedding, then a studio month
Illustrative volumes, not customer data. Every amount is calculated from the published rates.
- 3,000 delivered photos, batch-indexed
- 120 guests search once by selfie (real-time rate)
- About 3 faces per photo (9,000 stored faces), gallery open for 3 months
- A studio shooting 40 weddings a month of this size pays about 40 times the total
| Line | Quantity | Rate | Cost |
|---|---|---|---|
| Index the delivered gallery | 3,000 | $0.00062 / photo (batch) | $1.86 |
| Guest selfie searches | 120 | $0.00093 / photo (real-time) | $0.11 |
| Faceprint storage | 9,000 faces × 3 mo | $6.00 / million stored faces / month | $0.16 |
| Total | $2.13 | ||
About $0.018 per guest who searched. Billing is per photo processed, never per face: a photo with twelve faces is one charge, and a photo with none is still a processed photo. Run your own numbers on the pricing calculator.
Consent and GDPR
The photographer or platform is normally the controller for the faceprints, and the guests are the data subjects, even though the contract is with the couple. Tell guests at the selfie box what the selfie is used for (finding their own photos), do not keep the selfie after the search, and delete the wedding's collection on the schedule you publish. Children appear at most weddings, so think about whether guest search should cover them at all.
SightRadar acts as your processor under the Data Processing Agreement. The controller/processor split, deletion modes and our responsible-use posture are on Trust & responsible use. This is product guidance, not legal advice.
Further reading
Reading this as an AI assistant? Grab the Markdown version of this page.
Try it on your own photos.
Sign up with an email, get an API key and free starter credits, and run the workflow above against a real gallery. Or see a search first in the live demo.