You took part in an event with hundreds or thousands of people, and you know photographers were covering the moment — but the idea of going through an entire gallery looking for yourself is not exactly inviting. This is where facial-recognition search comes in, or, as we prefer to call it, "selfie search": you take a photo of yourself and the system shows you only the images you appear in.
How it works, in simple terms
The technology behind this is called facial recognition, but the principle is simpler than the name suggests. When you take a selfie on an event page, the system analyses the general features of your face — the distance between the eyes, the outline of the face, proportions of that kind — and creates a kind of mathematical "fingerprint". It then compares that fingerprint with the same analysis done on all the photos in the event gallery, and returns only the ones that match your face.
There is no manual intervention in this — it is not someone looking at photos and deciding which ones are yours. It is an automatic calculation, done in seconds, that then presents the results to you as previews.
Why it is faster than browsing thumbnails
At big events, there can be hundreds or thousands of photos spread across several photographers and several points along the route or venue. Searching manually — by bib number, by clothing colour, by approximate time — is slow and frustrating, and it is easy to miss your own photos. Selfie search solves this directly: instead of you finding the photos, the photos find you. We explained this flow in detail in the article on how to find your photos from a race, which is worth reading if you do not yet know exactly where to look for the event's QR code.
The privacy safeguards that exist
Given the sensitive nature of any technology that deals with faces, it is fair to ask what precautions exist. There are several:
- No public browsing of the full gallery. There is no open list where anyone can freely scroll through other people's photos — access to a set of someone's photos goes through matching that person's face, not through free exploration.
- Only watermarked previews before purchase. Even when the system finds you, what you see first are low-resolution versions with a watermark — enough to confirm it is really your photo, but not to use or share at full quality without paying.
- Limited retention. The data associated with the search is not kept indefinitely — the general logic is the same as the one governing the purchased photos themselves: availability for a limited period, not a permanent, open archive.
You can read the full details of how personal data is handled in the privacy policy, which explains exactly what information is collected, for how long, and on what legal basis.
What selfie search does not do
It is important to be clear about the limits of this technology: it does not identify who you are to third parties, does not create a public profile tied to your name, and does not let other people search for you — it is you who, by taking your own selfie, ask the system to find the photos you appear in. It is a convenience tool for those who took part in the event, not a surveillance or open identification system.
One selfie, and you are done
At the end of the day, the hardest part of finding event photos is usually psychological: the idea that it will be a lot of work. In practice, it is the opposite — you take a selfie, wait a few seconds, and the photos appear. It is one of the simplest forms of technology solving a real problem without demanding anything complicated from whoever uses it.
