About Where Is This Place

Where Is This Place answers a single, universal question: where was this photo taken? It offers four focused tools — an AI-assisted photo location finder, a metadata viewer, a capture-date finder and a metadata remover — plus a practical guide that teaches the method behind each one.

The site is deliberately narrow. We do not trace people, run face recognition or build dossiers. We help you read what a photo file honestly contains, estimate what a scene honestly suggests, and clean up before you share. The question 'where is this place' deserves a straight answer with labelled evidence, and that is the whole product.

Photo geolocation has exactly two evidence sources, and we keep them strictly separated in every result. The first is metadata: cameras write EXIF data — GPS coordinates, timestamps, camera model — into image files. When that data exists, the answer is exact and free: our tools read it locally in your browser, and the file never touches a server.

The second is visual estimation: when a file carries no location, an AI model analyses the scene itself — architecture, vegetation, road markings, signage — and ranks the most likely places. This is genuinely useful and genuinely imperfect. Landmark photos often resolve within kilometres; a plain room may be unidentifiable by any model on earth. Every AI result on this site is labelled an estimate with its confidence tier, because a confident-sounding wrong answer is worse than an honest 'we could not tell'.

The pipeline always runs metadata first, AI second. It is faster, cheaper, more private, and more exact whenever the file itself carries the answer.

Three of the four tools are fully local: the metadata viewer, the date finder and the metadata remover parse your file inside your browser tab with an open-source engine, and no image, filename or coordinate ever leaves your device. That is not a promise about behaviour — it is how the code is built; you can watch the network tab stay silent.

The location finder is local first too. Only when a photo has no GPS data does the browser build a resized, metadata-free copy and send that single copy to a visual geolocation provider through our server, which proxies the request so the provider key is never exposed. Originals are never transmitted, nothing is stored, and the analysis copy is discarded when the request ends.

This site is an independent tool. It is not affiliated with, endorsed by or connected to Google, OpenStreetMap, Geoapify, GeoSeer, Picarta or any other provider whose technology appears in the stack. Map tiles and geocoding come from third-party services under their own terms, and their names appear only to credit them honestly.

We are also not a verification authority. Metadata can be edited; AI estimates can be wrong. Our job is to give you labelled, sourced results and the means to verify them yourself — not to certify facts about the world.

The site runs on Next.js with the EXIF engine built on exifr, the open-source parser widely used across the imaging industry, extended with our own WebP container reader. Metadata removal uses lossless segment stripping with an automatic verification re-scan, and the whole interface works in six languages: English, French, German, Spanish, Japanese and Korean.

Everything user-facing is free and requires no account. There are no tracking pixels on tool pages, no upload logs with filenames, and no analytics events that record coordinates or EXIF content.

Found a wrong claim, a translation glitch or a tool that misbehaves? Write to support@whereisthisplace.world — corrections from users are how small tools stay honest. If you are reporting a location result, include the general context (country, scene type), never the photo of a private individual.

The metadata path: EXIF, the Exchangeable Image File Format defined by JEITA/CIPA, stores GPS coordinates as rational numbers in a dedicated IFD block. We parse that block with exifr, validate the coordinates against hard limits (latitude ±90, longitude ±180) and reject the placeholder (0,0) known as Null Island instead of mapping it. A location that survives those checks is displayed as 'found in photo metadata' with its exact source field.

The AI path: when no usable GPS exists, the browser draws the image to a canvas at a maximum long edge of 2048 pixels and re-encodes it as JPEG — a derivative that carries no metadata by construction. That copy, and only that copy, is analysed by the visual geolocation provider, which returns up to three candidates with confidence values. We translate those values into plain confidence tiers and always present them as estimates.

Why this order matters: metadata reading is instant, free, exact when present, and completely private. Spending an AI call on a photo that already contains its answer would waste money, time and privacy for a result that is usually worse than the stored coordinates. Tools that skip the metadata step are not simplifying — they are guessing when they could know.

Every page is written and maintained in English, French, German, Spanish, Japanese and Korean — localized, not machine-translated line by line. Each language version carries its own localized examples and phrasing, because 'where is this place' is asked differently in each market, and a tool that speaks your language should also think in it.

If a translation reads oddly, tell us: translation corrections are among the messages we action fastest.

Who runs it? A small independent team; there is no outside investor steering the tool choices you see. How is it free? The three local tools cost almost nothing to run because your browser does the work; the AI analysis is rate-limited per session and protected against abuse. Do you keep my photos? No — originals never leave your device, and analysis copies are discarded with the request.