Find someone from a photo

A face search is not a reverse image search

The two are easy to confuse, and the difference is the whole point of this page. A reverse image search — the kind Google and Yandex do for free — looks for the same file. It is good at finding where one exact picture has been republished. Crop that picture, filter it, or hand it a different photo of the same person, and it comes back empty.

A face search works the other way. It reads the geometry of the face — the fixed proportions between the eyes, nose and mouth that a haircut, a filter or five years do not change — and looks for that same face on any photo, not only the one you uploaded. That is why it can find a person from a picture that has never been posted anywhere: it is not matching the file, it is matching the face.

Start with the free routes

Before you pay for anything, spend a few minutes on the tools that cost nothing. Often they are enough, and we would rather you find your answer there than pay us for one you did not need.

Run the photo through Google Images and Google Lens first, then through Yandex. Yandex is unusually strong at faces — it can sometimes surface a different photo of the same person, closer to a face search than a plain file match. Crop tight to the one face before you upload; less background means a cleaner result.

• Google Images (images.google.com) — click the camera, upload the photo. Best at finding the exact same file republished somewhere else.
• Google Lens (lens.google.com) — reads the picture rather than the filename, and does better with a different, similar shot.
• Yandex (yandex.com/images) — the same camera icon, and the strongest of the free tools at faces; its own crop tool lets you isolate the one face before it searches. A paid search is only worth it once these come back empty.

What our face search does

Our search reads the geometry of the face in your photo — the fixed proportions that a haircut, a filter or the years do not move — and reduces it to a signature. It is that signature, not your photo, that gets compared, and it is the face it looks for: not the name, not the caption, not the handle.

What comes back is a list of public pages carrying that same face, strongest and newest first, each with a direct link you can open and a confidence score for how closely the two faces match.

When you don't know where to look

Every platform has its own search, but they all work from a name or a handle, and they assume you already know which one to open. If all you have is a face — no name, no username, no idea whether the person is on Instagram or LinkedIn or a forum you have never heard of — those searches have nothing to take hold of. You would be left opening each site in turn and comparing faces by eye.

This is the case a face search is built for. One upload runs the same face across the public web at once — the public profiles on Instagram, TikTok, Facebook, X, LinkedIn, YouTube, Reddit and Telegram, next to news sites, forums and blogs — so you do not have to pick the right platform first. If the face sits on a public page anywhere we index, it comes back in a single list.

Choosing the photo

The result depends almost entirely on the photo you start with. A clear, front-facing shot gives the search the most to work with; there is no trick that recovers a face the camera barely caught.

• Pick the sharpest, most front-facing photo you have. Eyes open, face unobstructed, good light.
• One face per search. If it is a group shot or a screenshot with several people, upload it as it is and crop to the one face you want on the next screen.
• Sunglasses, masks and heavy filters work against you. A plain, unedited face is the easiest to match.
• If you have several photos, try more than one — a different angle or a different year can surface a page the first one missed.

What you get back

A report of the public pages where the same face appears, each with a thumbnail of the matched image, a direct link, and a confidence score from 0 to 100. The report stays in your account, so you can return to it and open the links in your own time.

Read the score as similarity, not identity. A high number means two images are a close facial match; on its own it does not prove they are the same person. Two people can simply look alike, and relatives look more alike than most. Open the page, read the name, the dates, the other photos, and decide for yourself. We give you the source to check, not an answer to trust.

The honest limits

A face search is powerful, and it is not magic. Knowing where it stops is what keeps a result honest.

• It only sees public pages. Anything behind a login — a private account, a friends-only post, a closed group, a direct message — is invisible to us and to every public index. A face search cannot open a locked door.
• Someone who has never posted their own face will not appear, and plenty of real people never have. Their absence from the results is not evidence of anything.
• An account with no face on it — a pseudonym, a logo, a landscape — cannot be matched by face, however public it is.
• A match is a lead, not proof, and a clean result cuts both ways. Finding nothing does not make someone honest; finding a page does not make them who it says. Absence is not innocence; presence is not guilt.

What this is for

There is a right use for this and a wrong one, and the difference is not subtle. A face search is for verifying someone you are actually dealing with — the match who will not video-call, the stranger whose photo landed in your messages, the seller you have never met — or for finding where your own face has been posted without you. It is not for following someone who has not invited you into their life. Use it to answer a question you have a real stake in, and stop there. Every search is logged.