---
title: "About metadata.video: a private metadata inspector"
description: "Why metadata.video exists, how it reads, explains, edits and strips video and photo metadata on your own device, and what it promises: no uploads, no account."
url: https://metadata.video/about
---

# About metadata.video

Every video and photo carries a second, invisible layer of information: when it was recorded and in which time zone, very often exactly where, which phone or camera made it, the software it ran, and how the picture and sound were encoded. Most people never see that layer until it matters: a clip that reveals a home address, a holiday video that sorts itself into the wrong year, a file that won’t play on a TV for reasons nobody explains.

The tools that show this layer tend to fall into two camps. Expert tools list hundreds of raw fields with no explanation. Simple apps show four or five fields and hide the rest. Many online checkers ask you to upload the file first, which is an odd thing to do with something private. metadata.video was built to sit in between: everything, explained, and never leaving your device.

## What it does

* **Explains the file in a sentence**, then section by section: location on a map, a timeline of every date, the device, the codec and its compatibility, resolution and frame rate, bitrate over time, HDR and colour, audio channels, tracks and chapters.
* **Shows what the file gives away** in the Privacy section, rated by how much it matters.
* **Edits the metadata**: title, description, date and time zone, location, device, rotation flag, 360° flag, cover art, chapters and track names.
* **Removes it**, all of it or only the location, and then reads the new file back to show exactly what changed.
* **Describes the content** with optional on-device analysis: scene changes, colours, sharpness, loudness, face counts and speech.
* **Exports a report** as text, JSON or CSV, with file hashes for your records.

## How it works

The site is a set of static pages; the inspector runs in your browser. Three readers look at each file: our own reader for MP4 and MOV boxes and for photo EXIF, XMP and IPTC; Mediabunny, an open-source media library built on the browser’s WebCodecs; and MediaInfo, the long-standing open-source analyser, compiled to WebAssembly and served from this site. Bitrate graphs and frame timing come from the size and timing of every frame, read without decoding the picture.

Editing an MP4 or MOV rewrites only the small metadata boxes and copies the picture and sound byte for byte, so it takes a moment even for files many gigabytes in size. Other containers, such as MKV and WebM, are copied stream by stream into a new file with ffmpeg compiled to WebAssembly, again without re-encoding. JPEG, PNG and WebP photos keep their compressed image data untouched.

## Promises

* Your files are never uploaded. There is no server that could receive them.
* Free, with no watermark, no sign-up and no limit on file size beyond what your device can handle.
* No advertising. Anonymous visit statistics, which you can turn off in **Cookie settings**.
* Honest limits: metadata can’t prove a file is genuine, can’t reveal whether pixels were made by AI, and removing it doesn’t hide what is visible in the picture.

## Who it’s for

People about to share a video or photo who want to know what goes with it; editors and archivists fixing dates and titles; anyone puzzled by “codec not supported”; journalists and researchers checking where a file came from; and developers who want a MediaInfo-style report without installing anything. metadata.video is an independent, self-funded project.